EXTRACTION, NUTRITIVE VALUE, AND ACCEPTABILITY 0F ALFALFA LEAF PROTEIN FRACTIONS Thesis for the Degree of M. S.‘ MICHIGAN STATE UNIVERSITY RUTH PORTEOUS HAMILTON 1971 ..... ...... . argqr ABSTRACT EXTRACTION, NUTRITIVE VALUE, AND ACCEPTABILITY 0F ALFALFA LEAF PROTEIN FRACTIONS BY Ruth Porteous Hamilton One of the novel supplements which has shown potential in allevi- ating world protein shortages is leaf protein. 'Unfortunately, when extracted by classical methods it possesses an undesirable green color and grassy flavor. Consequently, this study, using alfalfa, was under- taken to investigate a simple method of extraction which might yield a light colored and relatively bland protein isolate with good biological value and organoleptic acceptability. Potentials of the by-product mother liquor were also investigated. Basically, the extraction procedure involved blending fresh alfalfa with water, pressing the juice from the pulp, removing the chloroplasts by mild heating and centrifugation, increasing the pH to 8.5 followed by centrifugation, and acidifying to pH 4.5 followed by centrifugation. This yielded the acid precipitate and acid supernatant. The protein isolate (acid precipitate) and liquor (acid supernatant) extracted from three cuttings of field alfalfa made during the summer were analyzed for proximate composition, amino acid composition, nutrit- ive value (using meadow voles), and acceptability. With young and green alfalfa 60% of the nitrogen in the leaf was extracted and of the extracted nitrogen 27% appeared in the precipitate Ruth Porteous Hamilton (or 3.3 g dry protein/lb wet alfalfa) and 30% in the supernatant (or 3.7 g/lb), the remainder being a part of the chloroplastic and alkali precipitates. The precipitate contained about 72% protein, whereas the supernatant contained only about 25% protein. Both had desirable amino acid distributions, being high in threonine, lysine, and tryptophan, but limiting in methionine. The nutritive value of the precipitates varied slightly with cutting, that of the precipitates from last two cuttings close to that of casein. The supernatant fractions (shown to have high saponin levels), particularly the last two cuttings, were toxic to the meadow vole. Necropsy indicated the primary lesion to be enteritis. Organoleptically the precipitate--fairly bland and light colored-- was acceptable, the supernatant--bitter and brown-dwas not. I 'II xllIll‘rI\ int-IIIII‘ [{,"|L.(I\I/I. EXTRACTION, NUTRITIVE VALUE, AND ACCEPTABILITY OF ALFALFA LEAF PROTEIN FRACTIONS By Ruth Porteous Hamilton A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1971 ACKNOWLEDGEMENTS I wish to say thank you to Dr. J. R. Brunner for his timely advice and for the independence he allowed me in this investigation. Particu- larly, I have appreciated his tolerance, good naturedness, and suggestions in the writing of this thesis. Thanks are due my husband, Tom, for keeping me going during some long hours of research and for his needed assistance in the preparation of this manuscript. I must also extend appreciation to Miss Ursula Koch for assistance with the amino acid analyses; to Dr. F. C. Elliott and Miss Victoria Marcarian of the Department of Crop and Soil Sciences for performing the saponin analyses, providing the meadow voles and alfalfa necessary for this study, plus a great deal of advice and assistance concerning the biological assays; and to Dr. V. L. Sanger and Dr. D. L. Whitenack of the Department of Pathology for assistance with the vole necropsies. I sincerely appreciate the patience and tolerance of Dr. L. R. Dugan, Jr., and Dr. J. W. Thomas in reading this manuscript and their helpful criticisms and suggestions. I also wish to thank the Institute of Nutrition (NIH Training Grant GMD 1818) and the Department of Food Science and Human Nutrition for financial support during this study and for providing necessary research facilities. 11 TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . . . . . . . 3 Extraction. . . . . . . . . . . . . . . . . . . . . . . . . 4 MethOdS O O O O O O O 0 O O O O O O O O O O O O O O O 4 Factors Affecting Extraction . . . . . . . . . . . . 18 Yields 0 O O O O I O O O O O O O O O O I O O O O O O 22 quality and Nutritive value 0 O O O O O O O O O O O O O O O 24 Composition. . . . . . . . . . . . . . . . . . . . . 24 Gross . . . . . . . . . . . . . . . . . . . . 24 Amino acid. . . . . . . . . . . . . . . . . . 27 Biological Assays. . . . . . . . . . . . . . . . . . 34 In vitro Digestion Methods . . . . . . . . . . . . . 43 Effects of Storage . . . . . . . . . . . . . . . . . 45 Indigenous Toxic Substances. . . . . . . . . . . . . 47 Acceptability . . . 53 Potential 0 O O 0 O O O O O O C O O O O O O O O O O O O O O 58 EXPERIMNTAI‘ O O 0 O O O O O O I O O O O O I O O O O O O O O O O O 63 Chemicals and Materials . . . . . . . . . . . . . . . . . . 63 Chemicals. . . . . . . . . . . . . . . . . . . . . . 63 Alfalfa. C O O O O O O O O O O O O O O O O O O O O O 63 Preparative Procedures. . . . . . . . . . . . . . . . . . . 64 Standard Extraction Method, Modifications A, B, C, D . . . . . . . . . . . . . . . . . . . . . 64 Modification E Modification F . Modification G . . . . . . . . . . . . . . . . . . . 74 Modification H Mbdification I Yield Study and Pirie Extraction . . . . . . . . . . 75 Field Hat-vests O O O O O O O O O O O O O O O O O O O 79 . . . . . . . . . . . . . . . . . . . 74 Chemical Methods. . . . . . . . . . . . . . . . . . . . . . 80 iii Page Nitrogen . . . . . . . . . . . . . . . . . . . . . . 80 Nonprotein Nitrogen. . . . . . . . . . . . . . . . . 81 Amino Acids. . . . . . . . . . . . . . . . . . . . . 81 Methionine and Cystine . . . . . . . . . . . . . . . 83 Tryptophan . . . . . . . . . . . . . . . . . . . . . 84 Carbohydrate . . . . . . . . . . . . . . . . . . . . 85 Lipid. . . . . . . . . . . . . . . . . . . . . . . . 86 Ash. . . . . . . . . . . . . . . . . . . . . . . . . 86 Moisture . . . . . . . . . . . . . . . . . . . . . . 86 Saponin. . . . . . . . . . . . . . . . . . . . . . . 87 Physical Methods 0 O O O O O O O O O I O O O O O O O O O O O 87 High-Voltage Paper—Electrophoresis . . . . . . . . . 87 Biological Methods 0 O O O O O O O O O O O O O O O O O O O O 88 Protein Efficiency Ratio . . . . . . . . . . . . . . 88 Composition of diets. . . . . . . . . . . . . 88 Test animals. . . . . . . . . . . . . . . . . ‘ 91 Experimental procedure. . . . . . . . . . . . 91 Post—experimental autopsies. . . . . . . . . . . . . 92 Organoleptic Evaluation . . . . . . . . . . . . . . . . . . 92 Cookie Formulation . . . . . . . . . . . . . . . . . 92 Taste Panel C O O O O O O I O O O O O O O O O O O O O 93 Statistical Analysis of Data . . . . . . . . . . . . 93 RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 96 Extraction. . . . . . . . . . . . . . . . . . . . . . . . . 96 Method . . . . . . . . . . . . . . . . . . . . . . . 96 Characteristics of Fractions . . . . . . . . . . . . 98 Solubility. . . . . . . . . . . . . . . . . . 98 Color and flavor. . . . . . . . . . . . . . . 99 Proteolytic Activity in the Juice. . . . . . . . . . 102 Enzyme Extractions . . . . . . . . . . . . . . . . . 106 Cellulase . . . . . . . . . . . . . . . . . . 106 Amylase . . . . . . . . . . . . . . . . . . . 109 Pectinase . . . . . . . . . . . . . . . . . . 109 Yields . . . . . . . . . . . . . . . . . . . . . . . 111 Greenhouse alfalfa. . . . . . . . . . . . . . 111 Field alfalfa . . . . . . . . . . . . . . . . 117 Total solids. . . . . . . . . . . . . . . . . 126 Summary . . . . . . . . . . . . . . . . . . . 127 Quality and Nutritive Value . . . . . . . . . . . . . . . . 130 Chemical Evaluation. . . . . . . . . . . . . . . . . 130 Gross composition . . . . . . . . . . . . . . 130 Amino acid composition. . . . . . . . . . . . 134 Biological Evaluation. . . . . . . . . . . . . . . . 144 Protein efficiency ratio. . . . . . . . . . . 144 iv Page Toxicity. . . . . . . . . . . . . . . . . . . 149 Acceptability . . . . . . . . . . . . . . . . . . . . . . . 153 CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 APPENDIX 0 o o o o o o o o o o o o o o o o o o o o o o o o o o o o 181 Table 10 11 12 13 14 15 16 LIST OF TABLES Page Compositions of alfalfa fractions obtained by the impulse rendering process . . . . . . . . . . . . . . . . . . . . . 24 Compositions of alfalfa and alfalfa juice before and after extraction with ethanol . . . . . . . . . . . . . . . . . . 25 Proximate composition of products from wet fractionation. . 25 Compositions of some fractions obtained from the extraction Of lucerne. O I O O O O O O I I O O O O O O O O O I O O O O 26 Amino acid composition of several alfalfa leaf protein preparations 0 O O O O O O O O O O O O O O O O C O O O O O O 30 The essential amino acid composition of protein from leaves, corn, soybeans, meats, and milk . . . . . . . . . . . . . . 33 Mean values for the retention of nitrogen by malnourished infants fed milk (M) or leaf protein and milk (LPM) . . . . 38 Mean values for nitrogen absorption and retention by mal- .nourished infants fed a milk diet (DSM) and a leaf pro- tein and milk diet (LP) . . . . . . . . . . . . . . . . . . 39 Nutritive value of lucerne leaf juice, LPG, and the chloroplastic and cytoplasmic fractions of leaf protein . . 42 Mean hedonic ratings of non-LPC and LPG incorporated prOductS O O O O O O O O O O O O O O O O O O O O O O O O O O 56 Composition of vitamin diet fortification mixture . . . . . 89 Composition of salt mixture W . . . . . . . . . . . . . . . 89 Formulations of control and experimental diets for wean- 1 ing meadow vales O O O O O O O O O O O O C O O O O O O O O 90 Formulations of the control and experimental cookies for organoleptic evaluation . . . . . . . . . . . . . . . . . . 92 Characteristics of fractions prepared by Modification A . . 97 Characteristics of supernatant fractions before and after passage over an ion exchange bed. . . . . . . . . . . . . . 102 vi Table . Page 17 Changes of alfalfa juice NPN from 0-24 hr after extrac- tion 0 O O O O O O O O O O O O O O O I O O O O O O O I O O O 103 18 Nitrogen contents of fractions obtained after blending With cellzyme O O O O O O O O O O O O O O O O O I O O O O O 107 19 Nitrogen contents of fractions obtained after incubation of alfalfa juice with cellulase . . . . . . . . . . . . . . 108 20 Nitrogen contents of fractions obtained after incubation 0f alfalfa jUice With alpha-amYIaSe o o o o o o o o o o o o 110 21 Nitrogen contents of fractions obtained after incubation of alfalfa juice with pectinase . . . . . . . . . . . . . . 111 22 Compositional characteristics of alfalfa leaf fractions obtained by yield study conducted concomitantly with Modification A of the standard extraction method (Extrac- tion I) . . . . . . . . . . . . . . . . . . . . . . . . . . 112 23 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method (ExtraCtion II) I O O O O O O O O O O O O O O O O O O O O O 113 24 Compositional characteristics of alfalfa leaf fractions obtained by yield study conducted concomitantly with Modification B of the standard extraction method (Extrac- tion III) . . . . . . . . . . . . . . . . . . . . . . . . . 114 25 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method from the first harvest of the first cutting of field alfalfa . . . . . . . . . . . . . . . . . . . . . . . . . . 118 26 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method from the fifth harvest of the first cutting of field alfalfa . . . . . . . . . . . . . . . . . . . . . . . . . . 119 27 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method from the second harvest of the second cutting of field alfalfa . . . . . . . . . . . . . . . . . . . . . . . . . . 121 28 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method from the fifth harvest of the second cutting of field ‘ alfalfa . . . . . . . . . . . . . . . . . . . . . . . . . . 122 29 Compositional characteristics of alfalfa leaf fractions obtained by yield study of the standard extraction method from the third cutting of field alfalfa . . . . . . . . . . 125 30 Total solids characteristics of fractions extracted from alfalfa . . . . . . . . . . . . . . . . . . . . . . . . . . 127 vii Table Page 31 Nitrogen characteristics of the fractions obtained from several extractions of alfalfa. . . . . . . . . . . . . . . 128 32 Proximate compositions of alfalfa leaf protein fractions. . 131 33 Amino acid composition of alfalfa leaf protein fractions expressed as grams of amino acid/100 grams protein. . . . . 135 34 Chemical scores of alfalfa leaf protein fractions . . . . . 138 35 Percentage of each essential amino acid as compared to that in the 1957 FAO provisional pattern. . . . . . . . . . 140 36 Percentage of each essential amino acid as compared to that in egg . . . . . . . . . . . . . . . . . . . . . . . . 140 37 Percentage of each essential amino acid as compared to that in cow's milk. I O O O O O O O O I O O O I O O O O O O 141 38 Percentage of each essential amino acid as compared to that in human milk 0 O O O O O O O O O O O O O O O O O I O O 141 39 Protein efficiency ratios of alfalfa leaf protein frac- tions and casein. . . . . . . . . . . . . . . . . . . . . . 145 40 Feed intakes (in g/day) of voles. . . . . . . . . . . . . . 149 41 Mean values and statistical significance for organoleptic variables for the acid precipitate. . . . . . . . . . . . . 154 42 Mean values and statistical significance for organoleptic variables for the acid supernatant. . . . . . . . . . . . . 155 Al The principal chemicals used in this study and their sources 0 o o o o o o o o o o o o o o o o o o o o o o o o o 181 A2 Amino acid composition of alfalfa leaf protein fractions expressed as moles of amino acid/1000 moles amino acids . . 184 A3 Amino acid composition of alfalfa leaf protein fractions expressed as grams of amino acid/100 grams sample , , , , . 185 viii Figure 10 11 12 13 14 15 16 17 LIST OF FIGURES Product utilization distribution scheme . . . . . . . . Diagram of the impulse rendering process. . . . . . . . Flow sheet for wet fractionation of alfalfa . . . . . . Flow sheet for extraction of protein from leaves by reverse osmosis . . . . . . . . . . . . . . . . . . . . Flow diagram for standard extraction method . . . . . . Flow diagram for Modification A of the standard extrac- tion me thOd O O O O O O I O O O O O O O O O O O O O O 0 Flow diagram for Modification B of the standard extrac- tion me thOd O O O O O O O O O O O O O I O O O O O O O 0 Flow diagram for Modification C of the standard extrac- tion me thOd I O O O O O O O I O O O I O O O I O O O O 0 Flow diagram for Modification D of the standard extrac- tion me thod O O O O O O O O O O O O O C O O O O O O O 0 Flow diagram for Yield Study of standard extraction methOd. O I O O O O O O O O O O O O O O O O I O O O O O Extraction by modification of method of Morrison and Pirie (1961). . . . . . . . . . . . . . . . . . . . . Judging form for the organoleptic evaluation. . Acid precipitates and acid supernatants obtained by the standard extraction procedure from three cuttings . . . Acid precipitates obtained by Pirie extraction. . . . . Increase in NPN as a percentage of the total nitrogen content of unheated alfalfa juice . . . . . . . . . . . Increase in NPN as a percentage of the total nitrogen content of heated alfalfa juice . . . . . . . . . . . Schematic representation of high—voltage electrophoreto— gram showing peptide bands observed . . . . . . . . . . ix Page 10 12 15 16 66 68 69 71 73 77 78 94 100 100 104 104 106 I Figure 18 19 Page Changes in protein, carbohydrate, ash, and lipid contents of the acid precipitate from the first through the third cutting . . . . . . . . . . . . . . . . 133 Changes in protein, carbohydrate, ash, and lipid con- tents of the acid supernatant from the first through the third cutting O O O O O O O O O O O O O O O O O O O O 133 \r’ INTRODUCTION Work on the practical utilization of leaf protein* as a dietary protein supplement was initiated some thirty years ago. Equipment has been designed to extract the juice, with reasonably good yield, from several species of green plants. Classically, the protein is precipi— tated from the juice by either heat or acid. Filtration or centrifuga-. tion yields a green cake, termed leaf protein concentrate (LPC), which on a dry basis is generally about 60-70% protein, 20-30% lipid, and has small amounts of carbohydrate and ash. It has a satisfactory distribu- tion of amino acids with the exception of methionine; and limited biological assays have shown LPC to have a nutritive value greater than oilseeds but less than casein or egg. Problems have arisen concerning acceptability, for LPC imparts a dark green color and grassy flavor to food preparations. Also, the high lipid content (particularly of unsaturated fatty acids) severely restricts the shelf life of the product. These objections can be largely overcome by solvent extraction of the LPC. However, this is an expensive process. Therefore, it would be desirable if the extraction process could be varied slightly so that the final product would be lighter in color, blander in flavor, and contain a lower lipid content. Other desirable features would be to increase nutritive value and protein content. * Presently, leaf protein is understood to be that protein extracted from the green portion of the plant, i.e., to include both stem and leaves. 1 2 Therefore, investigations concerning several modifications of the extraction procedure were made. These were followed by: analyses of the protein precipitate for proximate composition (protein, carbohydrate, lipid, and ash) and amino acid composition, biological assays using the meadow vole, and an organoleptic evaluation. A great amount of speculation has been offered concerning the poten- tial of the by-product mother liquor, particularly as a liquid feed supplement. However, no systematic studies of this fraction have been published. Therefore, it was subjected to the same analyses for quality (by chemical and biological evaluations) and acceptability as was the protein precipitate. Alfalfa was chosen because of its widespread availability, ease of extraction, and high production of protein and essential amino acids per acre when compared to other common forage or seed crOps. This legume is readily available in the tropics where the need for protein supplements is greatest. LITERATURE REVIEW The critical shortage of food, particularly that with a high pro- tein content, has resulted in widespread malnutrition throughout the world. The consequences of failure to remedy this problem have been reiterated by numerous authors (Pirie, 1959c, l969a; Byers, Green, and Pirie, 1965; Borgstrom, 1967; President's Science Advisory Committee [PSAC], 1967; Freeman, 1968; Brown, 1968; Munro, 1968; Altshul, 1966, 1969; Oke, 1969). Proposed solutions are generally classified as con- ventional or nonconventional. Conventional solutions suggest improved management of areas already under cultivation; better utilization of irrigation, drainage, and fertilization; better use of available animal breeds and crop varieties; increased disease and pest control; further technical development; and more efficient marketing, processing, and distribution of food (Pirie, 1953, 1967b; PSAC, 1967; Freeman, 1968; Buchanan, l969a; Borlaug, 1971). However, most authors agree that the problem will not be solved without the use of some supplementary protein sources. Several sources which deserve investigation are soybean, cottonseed, peanut (groundnut), coconut, fish protein concentrate, microorganisms cultured from petroleum, synthetic nutrients, algae, and leaves (Byers et aZ., 1965; Parpia, 1967; Borgstrom, 1967; PSAC, 1967; Wilcke, 1968; Buchanan, 1969a; Altshul, 1969; Russo, 1969a,b; Pirie, 1967b, l969a,b,c; Anderson, 1971). One single source will apparently not solve the protein shortage problems. Therefore, as long as protein deficiency exists all potential sources must be explored and evaluated 3 4 (Oke, 1969; Pirie, 1969a,b,c; Kinsella, 1970). Green leaves are one candidate, and work on leaf protein is intended as a complement, not alternative, to work with other protein sources. Extraction Methods The high fiber content of leaf requires that, for human consumption, the protein be separated from the fiber. The extraction of protein from leaves was first reported by Rouelle in 1773, who precipitated the pro- tein of hemlock leaves. The first significant work was reported by Osborne, Wakeman, and Leavenworth (1921) and by Chibnall and Schuyver (1921). The first group extracted 44% of the leaf nitrogen from macerated alfalfa with a succession of water, alcohol, dilute alkali, and hot alkaline alcohol treatments. The second group treated disinte- grated leaf tissue with water saturated with ether, a cytolyzing agent, thereby obtaining a yield of slightly less than two thirds of the total leaf nitrogen. By this procedure reasonably pure proteins were extracted from spinach (Chibnall, 1924) and alfalfa (Chibnall and Nolan, 1924). Chibnall and his associates (Chibnall, 1922a,b,c; Chibnall and Grover, 1926; Chibnall et al., 1933) continued working toward improved labora- tory extraction procedures. In these early years, the work.was largely biochemical with little emphasis on the practical use of leaf protein. Ereky (1926) patented the first method for the large scale proces- sing of green leaves. He pulped the leaves with a large amount of water on a machine which was essentially a set of moving knives. Goodall (1936) used a modification of the grooved-roll sugar cane mill for extracting the leaf juice in one operation. Both procedures had been designed to extract nutrients other than protein. 5 Lugg (1939) macerated foliage in the presence of a mildly alkaline buffer. This acted as a protein dispersing agent, and thus extracted 92-95% of the total leaf nitrogen. Later, Lugg and Weller (1944) described the extraction of protein from pasture leaves with alkaline buffer containing lipoid solvents (ethanol and ether). These general procedures were modified by Crook (1946), who maintained mild alkalinity and a low salt concentration during grinding. Crook and Holden (1948) then reported the results of this process on 28 species of plants. 0n the average 75% of the total leaf nitrogen could be extracted. Smith and Agiza (1951) extracted with 90% formic acid and 95% ethanol and then, after adjusting the pH to 10.5, precipitated the protein by acid or heat. Jennings et al. (1968) found that leaf protein was readily extractable into a phenol-acetic aciddwater mixture. Slade (1937) was the first to consider the food potential of leaf protein which he designated as "green cheese." The following decade, World War II initiated fears of food shortages. This spurred interest in the possible use of leaf protein as a food supplement. Therefore, systematic work aimed at designing equipment for bulk leaf protein extraction was begun (Pirie, 1942a). A key proponent of the use of leaf protein then, as he is today, was Mr. N. W. Pirie of Rothamsted Experi- ment Station, Herts, United Kingdom. Classical studies of Pirie and his associates have led to the extraction of human consumable leaf protein on a P110t plant scale. Since World War II Pirie and his associates have continued this biochemical engineering research. Pirie (1953, 1967a) and Byers, Fairclough, and Pirie (1956a) described the separation of the protein from the fibrous and strongly flavored components of leaf as follows: (1) maceration and disintegra- tion of the leaf, (2) expression of the juice, and (3) precipitation of 6 the protein. The problems and requirements of large scale leaf processing have been discussed by Pirie (1952). Early work indicated that large scale addition of water was troublesome and complicated the coagulation step. Therefore, work.was initiated to design machinery to pulp freshly harvested crops without the addition of large amounts of water (Pirie, 1966a). The conventional screw expeller and meat mincer proved unsatis- factory because they compressed the charge--the fiber acting as an ultra- filter to trap the protein--and generated excess heat (Pirie, 1952, 1966a, 1967a; Davys, Pirie, and Street, 1969). Sugar-cane rolls, rod-mills, edge and end-runner mills, and doughbreakers were all undesirable in this application (Pirie, 1966a). It was concluded that single stage machines are inefficient because the operations of maceration and juice expression are distinct. Thus the system developed was a pulper and a press designed specifically to handle leafy material. The pulper, developed as a modification of the hammer mill, is essentially a drum with fixed beaters on a rotating axial shaft. One to two tons of wet crop can be handled per hour. The classical description of this extractor was given by Davys and Pirie (1960). The design and operation has also been dis- cussed in detail elseWhere (Byers et al., 1956a; Pirie, 1952, 1956b, 1957a, 1961a). Hammer mills, in which the material undergoes a succes- sion of impacts, have also been used by Smith (1940), Bickoff, Bevenue, and Williams (1947), Kohler and Graham (1951), and Raymond and Tilley (1956). The separation of the juice from the pulp, like the pulping operation, possesses problems unique to leaves and, therefore, requires a press with definite characteristics. These have been discussed by Pirie (1959b, 1966a). All practical methods depend on the application of pressure to a constrained mass of pulp. There are several methods by which pressure 7 can be applied (Pirie, 1959b). Details of the earlier cam operated presses have been published by Fairclough and Pirie (1954), Byers et al. (1956a), Pirie (1956b, 1957a, 1959b, 1961a), and Davys and Pirie (1960). Davys and Pirie (1965) discussed their large scale continuous press, a tensioned conveyor belt on which the fibrous pulp is fed and then pressed between the belt and a perforated pulley with no differential movement between the two. It has a capacity of 3 T fresh pulp per hr. A hydraulic press was also used by Bickoff et a2. (1947); whereas Smith (1940), Kohler and Grahm (1951) and Raymond and Tilley (1956) used a screw press for this same operation. For the non—industrialized community a simpler and smaller "Village Unit", which pulps and presses fresh leaves in one operation, has been constructed by Davys and Pirie (1963). This unit has a capacity of up to 500 lb of leaf per operation (1—2 hr). However, it is less efficient than commercial models due to insufficient disintegration of the leaf, and extracts only 40-50% of the available protein. Davys and Pirie (1969) recently described a laboratory scale pulper suitable for processing 2-3 kg samples of leaf. Both efficiency and yields are comparable to those of the large scale equipment. Davys et a2. (1969) described a complementary laboratory scale press, consisting of two vertical grooved platens, which handles 1 kg quantities of leaf pulp. This unit gives results similar to those obtained with the large press. Since its origin, the extraction equipment has undergone continued modification and improvement in both design and operational detail. This evolution is outlined in the Annual Reports of the Rothamsted Experiment Station (Fairclough and Pirie, 1954; Byers, Fairclough, and Pirie,1955; Pirie, 1959d, 1960, 1961C, 1962, 1963, 1964b, 1965, l966e, 1967c, 19698). 8 Starting with high quality crops, about one half of the protein appears in the first extract. And, if the fiber residue is repulped with the addition of water, about one half of the remaining protein can be extracted (Davys and Pirie, 1960; Morrison and Pirie, 1961). Currently, up to 90% of the nitrogen can be separated from as much as two tons of crop per hour (Arkcoll and Festenstein, 1971). Pirie has also designed a steel press which grinds small quantities of leaves by forcing them under extremely high pressures through a narrow slot of adjustable width. Oxygen can be removed from the system and the extraction carried out in the presence of an inert gas (Pirie, 1956a, 1959d, 1961b; Festenstein, 1961). This press has been improved slightly by McArthur and Miltimore (1964). Of the many possible methods for separating the protein from the juice, Morrison and Pirie (1961) found that sudden heating to 80 C by steam injection was the most convenient for large scale preparation and gave the most filterable curd. According to Pirie (l969d), this heat treatment does not damage the nutritional quality of the protein but inactivates the endogenous enzymes which cause undesirable changes in the stored product. In addition, partial pasteurization is achieved. However, heating does denature the protein and it becomes insoluble. The production of a curd by acidification also has this insolubility effect (Pirie, 1961a). Therefore, the functionality of leaf protein as an ingredient is somewhat limited in that partial hydrolysis is probably the only method for getting leaf protein into solution (Pirie, 1957b). Leaf curd can be formed by the above method in an automatic coagu- lation unit described by Byers et al. (1955) and Pirie (1957a). However, this unit is not satisfactory for use with the batch extractor. There- fore, Younus Ahmed and Singh (1969) designed one for use on a smaller scale. 9 On a laboratory scale coagulation can be realized by acidification to pH 3 or 4. The protein coagulum is filtered via a pump-fed filter prss. The resulting cake is crumbled, suspended, and washed in 10 to 20 times its volume of water by vigorous mechanical stirring, which removes undesirable, strongly flavored substances. Lowering the pH to 3.5 or 4.0 insures removal of alkaloids and improves the keeping quality of the final product. The suspension can then be refiltered and pressed (Mbrrison and Pirie, 1961; Byers et al., 1956a; Pirie, 1959c, 1969d, 1970). McDonald (1954) has shown that yield and crude protein content of the curd are dependent on the efficiency with which the mother liquor is removed from the wet protein coagulum. According to Singh (1969) the curd will contain variable protein and nonprotein constituents depending on the coagulation conditions and the nature of various constituents in the extract. The final product is a dark green cake (termed leaf protein concen- trate--LPC) containing 50-60% moisture and is only slightly soluble in neutral or alkaline water. On a dry basis it contains 9-11% nitrogen (about 50-68% protein), 5-10% starch, 5-30% lipid and 1-10% ash (Mbrrison and Pirie, 1961; Pirie, l966a,b, l969d, 1970). The consistency and texture are similar to cheese or pressed yeast (Byers et al., 1956a; Pirie, 1956b, 1959c, 1966a, 1967a, l969d; Mbrrison and Pirie, 1960, 1961; Buchanan, 1969a). The fibrous residue, containing about 30-50% of the total leaf pro- tein nitrogen, is usually about 1—3% nitrogen on a dry basis (Raymond and Tilley, 1956; Byers at aZ., 1965; Pirie, 1953, 1966c, 1967c). The composition of the liquor depends on the species and maturity of the plant. The following ranges have been reported: dry matter, 20-47 g/l; . [I’LE llll‘ {All}, \‘lllllx ifi'l .l‘I’fl)!‘ :rlri I‘l‘lllllirlllr ll." Il.Ivl|§ 1 10 nitrogen, 0.4-1.5 g/l; and carbohydrate, 4-40 g/l (Pirie, 1961a, 1967c). In summary, by the methods of Pirie and his associates, fresh green leaves are pulped and separated into juice (A) and fibrous residue (B) as shown in Figure 1 (Pirie, 1942a, 1952, 1953, 1956b, 1957a,b, 1961a, l966a,c,e, 1968, 1969d; Byers et aZ., 1965). Very little more than protein food for man and coagulum { fats } other nonruminants (A) containing starch juice which gives after amino acids stock drink, medium coagulation liquor amides for the growth of micro— containing sugars organisms, source of nucleotides low molecular weight salts, etc. compounds cellulose (B) most of the hemicellulose fodder (dry or ensilaged) fibrous lignen for ruminants, fuel, residue pectin substrate for microbial containing fermentation yielding protein methane for fuel and some of the {fats microbial proteins starch - Figure 1. Product utilization distribution scheme. speculation has been offered on the potentialities of the by-products of leaf protein extraction. Ferguson and Armitage (1944) reported experi- ments on the growth of bakers yeast on the mother liquor from bracken juice. Bickoff et a1. (1968) reported that an aqueous extract of fresh alfalfa was shown to stimulate the growth of microorganisms. Preliminary results of feeding trials with cattle suggested that the liquor may be a valuable additive to liquid supplements (Perry, Peterson, and Beeson, 1969). Studies by Oelshlegel, Schroeder, and Stahmann (1969b) showed that the fibrous residue produced a good silage which was readily eaten 11 by cows. The residue from lucerne* is unsuitable for most papermaking processes (Raymond and Tilley, 1956). Other British innovators, Slade, Branscombe, and McGowan (1945), employed the screw press to both shear the cells and to exert sufficient pressure to express the juice. Protein was coagulated by heat under slightly alkaline conditions. This produced a green curd devoid of fiber and containing about 50% protein. In a large scale operation the input was half a ton of grass per hour. This yielded about 33% of the total protein or about 25 1b crude protein per ton of fresh crop (Raymond and Tilley, 1956). Many workers followed the lead of Slade et al. in extracting the juice from fresh leaves with screw expellers of varying designs and modifications (Nebraska Chemurgy Project, 1947; Ayala and Johnson, 1951; Tallarico, 1952; Miller, 1952; Isajev, 1956). Powlings (Anon., 1953) patented his screw expeller--"Protessor"--for producing a protein concentrate from green vegetation that was suitable for feed- ing farmstock. It yielded 60-80 1b crude protein per ton of fresh crop (Raymond and Tilley, 1956). In the Netherlands, Deijs and Sprenger (1952) concluded that the Powling screw press was the most promising type expeller for large scale applications; and it has been used by Akinrele (1963) in Nigeria. Tilley et al. (1954) at the Grassland Research Institute in England used a modified Christy and Norris "coir sifter" and a screw expeller, sequentially, to separate the leaf juice from the fiber. Steam heat yielded a dark green curd and a clear brown liquor which were separated by a filter press. The curd (50% crude protein) was dried on a twin roll film dryer. Lucerne yielded 31% and *Both lucerne and alfalfa are common names for Medicago sativa. Alfalfa is the North American reference whereas lucerne is used in other regions of the world. This text will use that name found in the paper of the particular investigator being discussed. 12 23% of the crop crude protein as protein concentrate for the first and second cuttings, respectively. Chayen, another British worker, took a different approach. He used the impulse rendering (IR) process (see Figure 2) originally designed for the extraction of fat from cellular material (Chayen and Ashworth, 1953; Anon., 1959; Chayen et aZ., 1961; Smith, 1966). Protein is leaf ------ -b pass--1 ‘1: Impulse i i “Impulse Vibrating Renderer Renderer Screen protein water solution Washing fiber , Precipitation fiber Tank Screw'Squeezer 11’ I fiber (50% water) [K Centrifuges] Azeotropic Dryer ll 1} chlorophyll dry protein xanthophyll C3130 tene Figure 2. Diagram of the impulse rendering process. released in the IR process by a mechanical rupturing of leaf cells by rapidly delivered impulses generated by the rotating beaters of a modi- fied hammer mill and transmitted through a liquid medium (water or dilute solution of Na2C03 which may contain an antioxidant). The effluent is freed of fiber by passage over a vibrating screen. The protein is pre- cipitated by acidification, heated to 60-70 C, and centrifuged. The product is a green cake of about 80% moisture which may be subjected to azeotropic distillation with toluene or spray dried to yield an intensely green powder with a characteristic, grassy odor. This continuous process 13 with a throughput of one ton of leaves per hour can separate 75% to 85% of the nitrogen from good quality leaves (Anon., 1959; Chayen et aZ., 1961). The composition of the isolate can be varied by modifying the IR processing conditions (Smith, 1966). In 1943 Guha and his associates (Pal and Guha, 1953; Chakrovorthy and Cuba, 1960) made the first Indian attempt to obtain protein from leaves in a form suitable for human consumption. They elucidated condi- tions for maximum extraction of protein and special features of protein technology. More recently on a small scale, Guha (1960) extracted 60% of the protein from various indigenous leaves by maceration with 2% Na2C03 and precipitation by acidification and heat. A concerted effort to extract leaf protein is in progress at the Central Food Technological Research Institute (CFTRI), Mysore. Singh (1964), attempting to identify suitable plants for bulk extraction, used a meat mincer, hand squeezing, trichloroacetic acid precipitation and centrifugation to extract protein from several species of local plants. Using a batch extractor and heat precipitation, he followed this first study with an investigation of lucerne to standardize the conditions of extraction and processing (Singh, 1967). The batch extractability of the 24 plant species studied at CFTRI varied from 25-60% (Singh, 1969). Valli Devi, Rao, and Vijayarahaven (1965) used the simple method of Pirie (1957b) to extract protein from the leaves of eighteen species of plants. Also, on a laboratory scale, Sentheshanmuganathan and Durand (1969) extracted the protein from the leaves of the Ceylon rain forest by the following procedure: homogenization of the chopped leaves in a blender, filtration through cloth, increasing the pH, acidification, heat, filtration and acetone extraction. 14 In this country, most of the work on extraction processes has been done at the USDA Western Utilization Research Laboratory at Albany. The primary objective has been to make products useful as animal feeds. Products useable for human consumption have been only a secondary object- ive. However, Spencer at al. (1969) admitted that this latter objective could be the more important contribution of their wet processing technique. In this technique, freshly ch0pped alfalfa is treated with ammonia to raise the pH and the juice is expressed with sugar cane juicing rolls capable of exerting high pressures. This process, now named the PRO-KAN process, yields a pressed cake and a green juice devoid of fiber. Each of these fractions is further processed into the marketable commodities illustrated in Figure 3 (Kohler et al., 1968; Anon., 1969; Spencer et aZ., 1969; Browning, 1970; Knuckles et aZ., 1970). The alfalfa whey (b) contains amino acids, sugars, minerals, and vitamins and thus has poten- tial as an animal feed supplement. Fractions (c) and (e) could serve as poultry ration additives and (d) could be used for poultry, swine, or possibly as human food. Fraction (a), containing nearly 50% of the total protein, can meet standards for a good quality dehydrated alfalfa meal and would be suitable for ruminants. An evaluation of the use of pilot plant sugar cane rolls in alfalfa processing has been reported by Knuckles at al. (1970). Up to 33% of the total solids and almost 50% of the protein from fresh alfalfa can be extracted into the green juice. Recently, a pilot plant coagulation system capable of handling over 90 gal of juice per hr was evaluated (Spencer et al., 1971). Hartman, Akeson, Stahmann (1967), in an attempt to prevent the loss of water soluble nutrients and to solve the problems associated with drying the moist leaf protein cake, spray dried whole juice (alfalfa and 15 alfalfa ethoxyquin Pellet whole dehydrated mill alfalfa pellets 1' J‘ ethoxyquin‘ > Dehydrator 'fl Pellet leaf L Air leaf fraction mill pellets yr Vr separator A» stem fraction,|Pellet F$1ong pressed mill fiber cake il'_—'—-J pellets (a) 50% alfalfa alfalfa Evaporatoj?) solubles solubleéx (b) CoagulatorJ> Separator green cur PRO-KAN E) (protein- xanthophyll- chlorophyll ” concentrate protein concentrate éigSolvent extractor] (c) (d) xanthoph;Yl-carotene- chlorophyll concentrate (e) Figure 3. Flow sheet for wet fractionation of alfalfa. pea vine) immediately after extraction. This yielded a green powder containing 18.5% and 35% protein for the pea vine and alfalfa, respectively. Oelshlegel et a1. (l969a) noted that the spray dried juice will contain, in addition to soluble nutrients, toxic substances that are soluble in water or are bound to the protein. In Portugal a protein isolate is obtained from grasses by a process based on reverse osmosis (Browning, 1970). Consequently heat is not required and the resultant protein denaturation is avoided (see Figure 4). In Hungary, Tangl (1949) reported a method for extracting a protein concentrate from lucerne for pig and poultry rations; and in Russia, Parini (1953) developed machinery for extracting a protein paste from a wide range of green materials. 16 Harvester alkaline water _ Diffusion I ’7 tanks recycle ? I r_j Vibrating! liquid screen solids 1i uids LS i -bypass-- ' Press [ Filte_r}4.-‘EentrifugelL ‘r ~L r + 1 Boiler Reverse oil Ultrafiltration osmosis V unit cattle I Y food Dryer Refinery Dryer Reverse \ Fermentation 4} ll osmosis tank protein—carbohydrate \’ 1L {L extract protein protein- protein- isolate carbohydrate yeast Figure 4. Flow sheet for extraction of protein from leaves by reverse osmosis. Removal of the lipid solubles from the protein concentrate by solvent extraction results in a product lighter and more bland in flavor than the original cake. Slade et al. (1945) initiated decolorization by treat- ment with amyl alcohol. Pirie and associates found acetone convenient and efficient for the simultaneous removal of lipids and water (Byers et al., 1956b; Pirie, 1957b, 1958, 1959a; Morrison and Pirie, 1961). The product keeps "indefinitely" in air. Acetone extraction was also used by Valli Devi et al. (1965) and Singh (1969). Chayen et al. (1961) extracted with acetone followed by ether and with a mixture of hexane and ethanol. Hartman et al. (1967) employed a 95% ethanol extraction for spray dried juice. Kohler at al. (1968) compared the efficiency of acetone, methanol, normal propanol, and isoprOpanol and concluded that the latter two were able to remove most of the pigments with two extrac- tions. Poppe, Tobback, and Maes (1970) used successively water, ethanol, acetone and ether to yield a final product from lucerne which was a l7 colorless and tasteless powder. Simultaneously the protein content of the samples increased from 68% to 82-84%. The merits of solvent extraction have been debated. Pirie (1969d) feels that solvent extraction should be avoided because: (1) lipids, particularly the unsaturated fatty acids, are valuable dietary compon- ents and (2) the additional cost and technology would make leaf protein production prohibitive on a local basis. However, Buchanan (1968) has shown that solvent extraction may produce a significant improvement in the nutritive value of leaf protein preparations by increasing the digestibility of the concentrate. The Americans (Hartman at al., 1967; Kohler et al., 1968; Spencer et al., 1969) feel that solvent extraction is necessary to produce a product acceptable as human food. Freeze-drying gives a stable and light colored powdery product which will keep well in air. The texture of the product is related to the method of freezing. A soft fluffy product will result only if the freezing process takes less than 2-3 min. Details of this process have 'been given by Mbrrison and Pirie (1963) and Pirie (1964a). The presence of various carriers gave a desirable product (Pirie, 1959a). However, this process is considered too expensive to be economically feasible in developing nations (Morrison and Pirie, 1960; Kohler et al., 1968). In summary, the basic extraction process is: (1) maceration of the leaves, (2) press extraction of soluble proteins, (3) precipitation of the protein by heat or acid, and (4) elimination of water (Kinsella, 1970). The large scale machinery which has been developed for the extraction of the juice can be divided into four classes according to the method used for maceration: (1) moving knives, (2) hammer mills, (3) rollers, and (4) screw expellers. The first two require a press for separation of the juice. All have been designed for continuous operation 18 (Tilley and Raymond, 1957). Under the International Biological Program (IBP), machinery for protein extraction has been sent to India, New Guinea, Nigeria, Uganda, United States and Canada (Smith, 1969). As a result of considerable engineering development, several processes are available for extracting the protein from a variety of leaves with reasonably good yield. Factors Affecting Extraction Extractability (i.e., the per cent of protein extractable into the juice or concentrate relative to the total protein content of the leaf) is affected by a variety of factors other than species. Apparently, it depends largely on the physical state of the leaf. Young, tender, lush leaves extract much more easily, giving higher protein yields, than mature, dry leaves of the same species (Crook and Holden, 1948; Byers at al., 1956b; Byers, 1961; Festenstein, 1961; Mbrrison and Pirie, 1961; Byers and Sturrock, 1965; Pirie, 1942a, 1962, l969b; Singh, 1964, 1967; Spencer at al., 1969; Oelshlegel et al., 1969a; Knuckles et aZ., 1970). This can often be correlated with the greater initial water and nitrogen content of the former. Yield of extracted protein increases to a maximum when the dry matter is about 15% and diminishes as the dry matter increases and the plant matures (Pirie, 1964b). Byers and Sturrock (1965) stated that the protein nitrogen content decreased as the age of the plant increased. The suggestion was made that increased mechanical entrainment of the chloroplasts is associated with the decreased extractability (Pirie, 1960). Nazir and Shah (1966) agreed that more protein was extracted from young than from mature leaves although they found that neither total nitrogen nor dry matter were correlated with extractability. Byers and Sturrock (1965) also noted 19 that high extractability does not necessarily correlate with high nitrogen in the leaf. Young cereal leaves, even though they have a low nitrogen content, are easily extractable (Byers et aZ., 1956b). Singh (1969) has reported that extractability varies with the bio-pH of the pulped mass, maturity of the chloroplasts, and nature of other nonprotein constituents present as well as the previously mentioned factors. Unpublished results of Pirie indicated that the trichloroacetic acid (TCA)—induced precipitate of some wheat leaf extracts contained 10-20% more nitrogen than similar heat-induced precipitates (Singh, 1960). Singh (1960) compared protein precipitates prepared by four methods: heat, TCA, perchloric acid, and TCA with uranyl ions. The precipitate produced by TCA contained 10-15% more nitrogen than the precipitate produced by heat. He suggested that heat coagulation pro- motes either nuclease or protease activity. However, he has more recently stated that heat-induced coagulation yields a LPC with higher contents of dry matter and nitrogen than acid-induced precipitation (Singh, 1967, 1969). Sur (1967) could precipitate 96% of the juice pro— tein by lowering the pH of the juice to 3.5 before heating. Many researchers have noted that maceration in an alkaline solution or an increase in pH to about 8 prior to precipitation of the protein resulted in higher yields. Generally, they suggested that this was due to increased solubility of the protein (Lugg, 1939; Lugg and Weller, 1944; Crook, 1946; Smith and Agiza, 1951; Byers et aZ., 1956b; Chayen at al., 1961; Morrison and Pirie, 1961; Akinrele, 1963; Singh, 1964, 1969; Nazir and Shah, 1966; Pirie, 1966e, 1969b; Sentheshanmuganathan and Durand, 1969). Festenstein (1961) and Pirie (l959d) reported that the increased extraction under alkaline conditions was largely due to greater dispersion of the chloroplasts. It was noted by Spencer at al. 20 (1969), Browning (1970) and Knuckles at al. (1970) that the increase in pH in the PRO—XAN process also decreased the losses of xanthoPhyll, chlorophyll, and protein which result from enzymatic activity during rolling, and changed the consistency of the curd to one easier to process. However, Poppe at al. (1970) reported that although an increase in alkalinity increased yields, it decreased the purity (protein content) of the preparation. Substances such as soaps and surface active alcohols influence the complexing of proteins with other leaf constituents (Pirie, 1942a; Cllayeti et al., 1961), particularly chlorophyll and lipids (Festenstein, 1961). Smith (194la,b) and Smith and Pickels (1941) studied the effects of various detergents in clarifying leaf extracts. Re and Clendenning (1956) found that detergents may either free the pigment or leave it attached to lipoprotein prosthetic groups. Festenstein (1961; Pirie, 1959d, 1960), by using detergents to release nitrogen from the residual fiber, extracted 90-95% of the nitrogenous material in the leaf. Various workers reported proteolytic activity in leaf juice (c.f., Byers, unpublished; Pirie, 1942a). However, few substantiated their statements with convincing experimental evidence. Tracey (1948) reported that both sap and fiber of green leaves showed protease activity. He noted that the difficulty in working on the protease of green leaves is due to its low concentration. Singh (1962) observed the autolysis of protein in leaf extracts and the proteolytic activity of these extracts against casein when incubated at 37 C for 2 hr. Autolysis was directly related to the temperature and the time between extraction and precipi- tation. Nazir and Shah (1966) investigated the autolytic breakdown of leaf protein at 25 C and concluded that juice can be kept for 1-2 hr without an appreciable loss in protein content. Davys et al. (1969) 21 concluded that the most critical period during processing is between pulping and pressing because protein is lost both through proteolytic activity and protein coagulation on the fiber (Pirie, 1967c). A Characteristic property of plant leaf extracts is the almost immediate development of a brown color, due presumably to the action of polyphenoloxidase. Cohen et al. (1956) have shown that browning can be prevented when all operations during homogenization and fractionation are conducted in a nitrogen atmosphere. Under such conditions an increase in solubility of the protein and a reduction in the brown color of the extract was noted. Various workers have fractionated the leaf extract into (1) a chloroplastic protein and (2) a cytoplasmic or soluble protein. Festenstein (1961) obtained the two fractions, respectively, by centri- fugation at 1500 g for 15 min and then at 12,000 g for 15 min. Pirie (l969e) also employed centrifugation to achieve a similar classification. Singh (1967), by differential heat treatment-—heating at 50-55 C to obtain the chloroplastic proteins and further heating of the supernatant to about 80 C to precipitate the cytoplasmic proteins--found 40% of the total heat precipitable protein to be in the chloroplastic fraction (7% nitrogen) and 40-45% in the cytoplasmic fraction (ll-12% nitrogen). He believed that this process may produce a high protein fraction of light color with a flavor suitable for human consumption. According to Byers (1967b) and Pirie (1963, 1964b) the per cent chloroplastic to per cent cytoplasmic protein ratio varies with species, not maturity, although Henry and Ford (1965) claimed that there is a higher prOportion of cytoplasmic protein present in extracts from older leaves. 22 Yields In determining the desirability of a species, yield as well as ease of extractability must be considered. Stahmann, of the University of Wisconsin, has published valuable quantitative data concerning leaf pro- tein yields (Akeson and Stahmann, 1966; Stahmann, 1968a,b). He and his associates have calculated the yield of essential amino acids and total protein from various U.S. forage and seed crops which could be processed into LPC. The forage crops, particularly the legumes, generally pro- duced several times more protein and essential amino acids per acre than seed crops. Alfalfa gave the highest yields of essential amino acids (300 lb/acre) and total protein (2400 1b/acre) of all 25 cr0ps considered. Therefore, Stahmann suggested that it would be the best crap for the highest production of LPC per acre (Akeson and Stahmann, 1966; Stahmann, 1968a,b). The first reported yields (extracted protein/acre) from different crop leaves ranged from 26.2 lb for tares to 730 lb for sugar beet (Byers at aZ., 1957; Pirie, 1958). Byers and Sturrock (1965) reported yields of leaf protein obtained from the large scale extraction of several temperate crops and suggested that the use of a suitable succes— sion of craps should make it possible to obtain yields of 1000 kg protein/ hectare/yr. Protein yield appeared to depend on species, variety, season, physiological age of the plant, ability to regrow after cutting, and time of harvest. Since then, Pirie (l967a,c, 1969b,d) has reported yields to be between 1200 and 1400 kg extracted protein/hectare/yr. Byers and Sturrock (1965) and Oke (1967) both found cereals to give the highest yield of nitrogen. Regrowth patterns and the effects of different forms of husbandry on the yield of extractable protein have been studied at Rothamsted (Pirie, l966e). It was postulated I" \f1iltfl'll. Ia‘Il‘li... I I I.--) . h\||l.ll|.lull'l|\lr .. I f‘lllll..flb‘ll,.|‘ I'll f).|l)’ll‘r 1" 6r. ; 23 that the continual growth in the wet trOpics would probably produce yields more than twice those achieved in Britain (Pirie, 1966c, 1967a, 1968; Oke, 1969). Using estimates based on batch extraction yields of lucerne, Singh (1967) projected industrial yields to be 3-3.2 T extracted protein/ hectare/yr, which is quite exceptional for Indian agriculture (Pirie, 1969a,b,d). Actual yields of lucerne and of by—product vegetation grown on non-agricultural land have been 1500 and 66-160 kg protein/hectare/yr, respectively (Singh, 1969). Finally at Rothamsted, a yield of 2 T extracted protein/hectare/yr was realized with cereal crops which were followed by mustard or fodder radish (Pirie, l969d,e; Arkcoll and Festenstein, 1971). This was a considerable improvement over Slade's studies in which he reported 200 lb protein/acre of grass (Slade et al., 1945). In general, based on extractability and yield the legumes appear to be the best source of leaf protein (Buchanan, 1968). Although the per cent extractability tends to diminish with crop maturity, the total amount of protein per acre increases. This results in an optimal harvest time which depends upon species and cultural conditions (Pirie, 1961c, 1962, 1969b). Chayen et al. (1961) noted that the yield of protein falls when grass reaches the flowering stage. Various studies on alfalfa have been conducted to determine the optimal time of harvest in relation to nutrient content and total yields (Van Riper and Smith, 1959; Smith et al., 1966; Hanson et aZ., 1965; Ogden and Kehr, 1968; Smith, 1970). Fiber content increased with maturity while protein content and digestibility tended to decrease. However, the yield of total dry matter increased with maturity. All workers agreed that harvesting at 1/10 bloom is the best compromise to give ll I\1(|\u‘l , 'Inl| Ifi‘ It! I. f l. I >i\ ) I, \llllldrl‘r I. {fic (III)! ‘II 24 Optimal yield of high quality alfalfa protein from the vegetative plant. Quality and Nutritive Value Composition Cross. The curd from the extraction method of Slade at al. (1945) was reported to have the following composition: protein, 17.5%; chloro— phyll, 19.3%; carotene, 0.2%; xanthophyll, 0.3%; glycerides, 5.7%; wax, 3.4%; other lipid extractives, 18.0%; and ash, 5.6%. The composition of fractions derived from the impulse rendering of alfalfa have been reported by Smith (1966). Some of his data are given in Table 1. Table 1. Compositions of alfalfa fractions obtained by the impulse rendering process Fraction Total solids Crude protein Carbohydrate Lipid (Z) (Z) (Z) (Z) Alfalfa 20.0 22.0: -—-1 ———1 3 Lipid-protein isolate -—- 61.0 2 12.5 17.0 ’ Protein effluent ——- 0 01 -—- -__ 1Dry basis 2Wet basis 3Acetone-solubles Hartman et al. (1967) compared the proximate compositions of fresh green plant, spray dried juice, ethanol extracted spray dried juice and the fibrous residues to other high protein foodstuffs. Some of their data are given in Table 2. Juice from the second cutting appeared to have a lower protein content. 25 Table 2. Compositions of alfalfa and alfalfa juice before and after extraction with ethanol Fraction Moisture Protein Fat Ash Alfalfa (first cut) 5.5 21.7 --- --- (second cut) 5.1 19.5 --- --- Spray dried juice (first cut) 4.9 34.9 6.6 12.7 (second cut) 4.2 31.3 5.8 12.1 Ethanol extracted spray dried juice (first cut) 2.6 42.8 0.6 14.4 (second cut) 2.3 37.0 0.6 14.1 The compositions of the products from the USDA wet fractionation process are shown in Table 3. Table 3. Proximate composition of products from wet fractionation Fraction Moisture Protein1 Fat1 Ash1 (Z) (73) (%) (Z) Alf 2 3 alfa 79 23.4(22.5) --- —-- Juice 91 36.3(32.3) --- --- Protein-xanthophyll- 4 chlorOphyll concentrate 80 55.7(50.5) 6.7(8.1) 10.2(12.0) Alfalfa solubles 94 28.9(18.3) 0.5(0.6) 19.7(21.l) l'Dry basis; 2Figures not in parentheses reported by Spencer at al. (1969); Figures in parentheses for alfalfa and whole juice reported later by Knuckles at al. (1970); Figures in parentheses for PRO-XAN and solubles reported later by Spencer et al. (1971). Subba Rau, Mahadeviah, and Singh (1969) reported the composition of fractions extracted from lucerne. Some of their data are given in Table 4. The cytoplasmic fraction was low in mineral matter and had the highest protein content. Most of the mineral matter, the ether extract- ives and carbohydrate, and somewhat less of the total coaguable leaf protein was in the chloroplastic fraction. 26 Table 4. Compositions of some fractions obtained from the extraction of lucerne Fraction Mineral Ether Crude Carbohydrate matter1 extractives protein by difference (2) (Z) (Z) (Z) Batch I Lucerne --- —-- 24.0 --- Juice --- --- 28.0 --- Batch II LPC 6.0 11.5 61.0 19 5 Chloroplastic 11.0 13.5 46 0 26 0 Cytoplasmic 1.5 5.0 77 0 16 0 1Dry basis Oelshlegel at al. (1969a) reported the crude protein, crude fat, crude fiber, ash and nitrogen-free extract contents for fractions (start- ing material, residue, juice, protein precipitate) obtained during the extraction of LPG from green plant wastes. The composition of Pirie's product (already given) varied with the age and species of the leaf (Morrison and Pirie, 1961). The nucleic acid content depended on the details of the extraction procedure but was only a minor component due to rapid hydrolysis by leaf ribonuclease (Morrison and Pirie, 1961; Pirie, 1961a). The analyses of some protein concentrates extracted from fresh Indian flora by Valli Devi at al. (1965) showed wide variations in the protein (34-77%) and ash (1.3-9.0%) contents of the isolates. Other Indian workers, Sentheshanmuganathan and Durand (1969) observed a wide range in the protein content (27-75%) of the concentrate. Ash, calcium and iron were also determined. In the Philippines, Gonzales, Dimaunahan, and Banzon (1968) reported ranges in crude protein content of 9-39% and 9-63% for the leaf and protein isolate, respectively. And in West 27 Pakistan, Hussian, Ullah, and Ahmad (1968) found the crude protein con- tents of the leaf and the protein isolate to range from 17-30% and from 28-60%, respectively. The protein content of concentrates made from 58 tropical Species in Ghana varied from 15-70%; the legumes exhibiting the highest protein contents and greatest extractability (Byers, 1961). Eggum (1970) reported the composition--crude protein, ether extract, crude fiber, nitrogen-free extract, true protein, ash, Ca, P, Na, and Mg--of some leaves and LPG from eastern Nigeria. The crude protein con- tents were within a fairly narrow range of 31-37%. His results agree with those obtained by Oke (1966b) from leaf vegetables in Nigeria. Amino acid. Numerous amino acid analyses of leaf protein have been published. In 1939 Chibnall considered the essential amino acid compo- sition of leaf protein and concluded that it would have an excellent nutritive value. Lugg and his associates did a great deal of the early work in this area. In 1939, Lugg reported the contents of amide, tyro- sine (Tyr), tryptOphan (Trp), and the sulfur amino acids in the whole protein of leaves. He saw no evidence of compositional variation due to age, fertilization, climate or locality of growth; but a possible variance with species was observed. Lugg and Weller (1944) determined the amide, Tyr, Trp, cystine (Cys) and methionine (Met) nitrogen for the leaves of four pasture plants. Later, an estimation was made of the basic amino acids--arginine (Arg), lysine (Lys), and histidine (His) (Lugg and Weller, 1948a). It was determined that the composition of whole leaf protein from senescent tissue may vary markedly from that pre- vailing at an earlier stage of maturity (Lugg and Weller, 1948b). In 1949, Lugg reviewed the literature concerning plant proteins with regard to the amino acid composition and found little variation within species 28 due to conditions of growth and no outstanding differences between species. Armstrong (1951) determined the Tyr, Trp, Met, Cys, Lys, and His contents of fifteen herbage species by microbiological assay. He con- cluded that there may be a wider variation in amino acids of herbage species than previously observed by Lugg and co-workers. Smith and Agiza (1951) colorimetrically determined 14 amino acids (separated by two dimensional paper chromatography) for several species. They reported that the proportions of certain acids, particularly Arg, Lys, and glutamic (Glu),were affected by such factors as age, stage of growth, season, nitrogenous fertilizers and species. In 1953, Yemm and Folkes published data for the amino acids (determined chiefly by microbiological assay) for three fractions (total, cytoplasmic, and chloroplastic) of barley leaf protein. Their results showed a high degree of uniformity, with appreciable differences only for Lys. Reber and McVicar (1953) reported similar patterns of amino acid distribution in the bulk proteins of several cereal grasses with increases in glutamine (Gln) and isoleucine (Ile) and decreases in Lys in older plants. In contrast, others reported that the protein extracted from four species of grass at varying stages of growth showed considerable uniformity in Lys contents (Waite, Fensom, and Lovett, 1953). By micro- biological assay, Kelley and Baum (1953) concluded that leaf meals from several vegetables contained a well balanced mixture of the eight essen- tial amino acids plus His and Arg. The amino acid composition of amaranth leaves was determined by Deshpande and Rao (1954). Also that year, Kemble and MacPherson reported analyses of protein fractions extracted by several methods from a variety of herbages. Although the protein varied greatly in purity, there were 29 no significant differences in the monoamine and monocarboxylic acid con- tents. Except for Ile and alanine (Ala), their results are in good agree- ment with those obtained by Yemm and Folkes (1953). Kuppuswamy, Srinivasan, and Subramaniam (1958) noted changes in amino acid composition as the plants aged. The results of Pleshkov and Fowden (1959) confirmed the well known fact that total nitrogen and protein nitrogen decrease with age of the plant and vary with soil mineral treatments. The amino acid composition of barley leaf protein showed a trend toward decreased amide and Arg with slightly increased Lys contents in the older plants. The amino acid compositions of protein extracted from various leaves (lucerne, spinach, cabbage, and burr medic) exhibited little composi- tional variation and the concentrations for lucerne proteins are presented in Table 5 (Chibnall, Rees, and Lugg, 1963). Wilson and Tilley (1965) determined, by ion exchange chromatography, the amino acid compositions of protein isolated by alcohol precipitation from lucerne and grasses. They were similar and had no appreciable dif- ferences due to stage of growth (Table 5). The classical and most comprehensive set of amino acid analyses on leaf protein concentrates was published by Gerloff, Lima, and Stahmann (1965). Protein concentrates extracted from green leaves of ten species harvested under different conditions of fertilization and maturity by the mechanical methods of Morrison and Pirie (1961) and Chayen et a2. (1961) were analyzed by ion exchange chromatography. Although the pro— tein content ranged from 31.5-83.8% and was subject to crop condition, extraction procedure, and subsequent processing, the samples did not, with the exception of Lys, show large variation in amino acid content. The range of amino acid contents are given in Table 5 along with the Table 5. Amino acid composition of several alfalfa leaf protein preparations U H :8 001 um >sr-l 09-: IO '5 °'” "4H“ 0 5.455 (Ono-I on. H CO 4.) 0 ['4 \O 'U a8 (U U s 0) U U) H (U U a: m H '1'! In L!" (U ‘H H CO ‘H ‘H ‘H H o <: H H (D we 0) g M >4 0 a) O F+n 3 H00) HUN Elihu-l DO: E a o '3 (D U :4 .2 3 m a!“ .3 ‘H 5N H OH ‘31! H :g u (U 6.5 5.2 6.2 4.9 5.7 6.3 5.0 5.8 6.4 5.3 5.9 5.5 5.6 5.6 6.3 5.2 6.1 4.5-7.3 4.3-5.8 5.9-6.9 9.4 4.6 9.4 4.2 5.9 8.7 4.2 4.9 8.6 4.4 4.9 Lys Thr 5.6 5.1 Val com HO \OI-l Fir-I {x (fit-i v-lI-I \‘fm HI—l \‘TN HH Met Cys 5.0 10.3 4.9 5.3 5.0 9.6 4.6 6.0 9.8 4.3 4.9-6.6 3.2 4.4 5.1 Ile 3.5 6.5 Leu Tyr 7.0 4.8 8.6 4.7 8.0 4.7 9.1-10.7 3.1-6.1 7.5 3.1 7.5 2.8 6.7 4.9 5.0 2.8 2.3 30 QN <76] Phe Trp 2.0 3.0 1.9 2.4 2.5 2.3 6.2 10.2 1.3-2.9 5.0 15.7 5.1 16.6 4.1 5.2 16.5 13.0 His 6.6 9.1 8.1 9.4 7.3 9.7 6.7 11.3 5.9 13.8 5.8-7.3 Arg Asp 900-1008 3.8-5.4 11.0-13.2 8.0 7.8 4.2 6.5 4.1 6.5 4.8 10.0 4.9 4.5 10.6 11.4 5.2 12.0 4.8 11.3 3.9 4.5 Ser 806 4.0 7.7 4.5 7.0 3.2 7.3 Glu Pro Gly Ala 4.7 4.5 4.4 4.9 5.3 6.0 3.5-6.7 5.1 4.9 5.3-6.1 5.8-7.4 3.8 6.6 4.8 5.6 5.7 4.7 5.1 5.7 6.5 7.5 6.7 6.3 6.1 6.0 5.8 5.7 7.5 7.1 6.2 2 Extracted by ether method of Chibnall and Nolan (1924) 1 Average of two 44) pecies Determined separately. 5 ugg and Weller (19 of ten 3 7 4Range 0 ’ method of L 3 6Refers to cutting Extracted by maceration Expressed as per cent samples prepared by impulse rendering; 3 31 values determined for alfalfa. No significant effect could be identi- fied with species, variety or maturity. A comparison of the average essential amino acid values of LPC to those of animal proteins showed that only Lys and Met consistently were lower in leaf protein. When compared with other higher protein foodstuffs LPC is high in Met and Lys (Oke, 1969). With Met supplementation leaf protein was suggested to be a well balanced source of dietary protein (Gerloff et aZ., 1965). The amino acid composition of lucerne LPC reported by Singh (1967) is similar to that obtained at Wisconsin except for differences in leucine (Leu), valine (Val), and Arg. Hartman at al. (1967), in their analysis of spray dried alfalfa and pea vine juice, obtained amino acid compositions similar to those reported by Gerloff et al. (1965); see Table 5. With only two excep— tions the Met contents were as high or higher than those of the other protein supplements to which they were compared. On the basis of amino acid analyses, the authors concluded that leaf protein could have a nutritive value equal to or greater than the protein supplements cur- rently available except possibly fish meal. Unpublished results of Oelshlegel et a1. (Stahmann, 1968b) indicated that the distribution of the eight essential amino acids in alfalfa juice was similar to that recommended by FAQ. The amino acid compositions for alfalfa leaf protein, alfalfa leaf cytoplasmic fraction, and alfalfa leaf chloroplastic fraction obtained by the impulse rendering process have been reported by Smith (1966). All three fractions showed similar amino acid distribution. The con- tents of the essential amino acids are given in Table 5. The sulfur 32 amino acids are limiting.* However, small differences in their content do not seem to warrant the rather large variation given for the chemical scores--57, 65, and 54 for LPC, cytoplasmic and chloroplastic fractions, respectively. Valli Devi et a1. (1965) determined, by microbiological assay, the Met and Lys contents of LPG prepared from several Indian leaves to be l.3-2.2% and 3-4%, respectively. In contrast to such low values (rarely more than 2 g/l6 g N) reported for Met, Surinder Kaur and Vijayaraghavan (1961) found, in a survey of the Met contents of a number of protein concentrates prepared from various local sources, that in seven out of eight species a range of 2.2-2.7 g Met/16 g N existed. Amino acid analyses of various tropical leaf proteins by Sentheshanmuganathan and Durand (1969) indicated that, except for the sulfur containing amino acids, the essential amino acids were in amounts comparable with those in milk and soya flour and greater than those in the 1957 FAO provisional pattern. They calculated that the chemical scores (CS) (Mitchell and Block, 1964) of the various leaf protein iso- lates ranged from 50 to 73 against the 1957 FAO provisional pattern. Pirie (1962, 1969a) reported that there were small but definite dif- ferences between the amino acid compositions of the whole protein, the chloroplastic and the cyt0p1asmic fractions extracted at Mysore. The amino acid composition of some leaves and LPG, principally cassava, from Biafra have been reported by Eggum (1970). There was little variation between species and variety and between whole leaves and LPG. And once again the essential amino acid pattern was adequate except for Met. Shanley and Lewis (1969) determined the Lys, Trp, Cys, * Limiting amino acid: is that essential amino acid available in least amount in relation to its requirement (Bergen, 1971). 33 and Met contents of the leaves of a number of plants to estimate their value as a supplement to the maize meal staple diets of South Africa. In all cases the CS of the maize meal-leaf mixture (86) was much greater than that of maize meal alone (65). Table 6 (Stahmann, 1968a; Oke, 1969) lists the essential amino acid compositions of protein from green plants, maize, soya beans, meats, and milk. LPC had a higher content of Lys than maize, more Met than soya bean and compared favorably with the animal proteins. Leaf protein is a mixture of many different proteins from a meta- bolically active organ of the green plant. Therefore, large differences Table 6. The essential amino acid composition of protein from leaves, corn, soybeans, meats, and milk Protein source Lys Phe Met Thr Ile Leu Val Trp * LPC 6.3 6.0 2.1 5.2 9.8 5.3 6.3 1.6 Opaque 2 maize endosperm 3.6 4.5 2.1 3.7 10.5 3.8 5.7 --- Soya bean meal 6.4 4.8 0.6 3.7 3.5 6.1 5.0 1.2 Meat, poultry, fish 8 Milk 8. Ulb \I\O QDUI woo mu: #0.) Dow UIN H * Expressed as 3 amino acid per 100 g protein in the amino acid composition of protein from different species, or the same species in different physiological states, are not expected because it is unlikely that the same excess or deficiency would characterize most of the different component proteins (Singh, 1967; Oke, 1969; Pirie, 1957b, 1959c, l966a,c, 1969b; Morrison and Pirie, 1961). In agreement with this speculation, the observed differences in amino acid composi- tion with the possible exceptions of Lys and Met according to the studies 34 reviewed herein are generally small. Differences between species seem no more significant than those within a species. Methionine appears to be the amino acid which varies most. Jennings and Lewis (1969) demon— strated a high degree of Met loss during the hydrolysis of protein in leaf material which had not been previously oxidized. The range of Met loss was from 30-50%. The actual degree of destruction seemed to be unpredictable, as evidenced by poor correlation between some unoxidized duplicates. In summary, the various amino acid analyses suggest that leaf protein should be a better supplement than most of the cereal and seed proteins, but not as good as casein or egg (Pirie, 1959c, l969b,d). \ Biological Assays While amino acid analyses and resulting chemical scores provide an indication of protein quality, they give no evaluation of availability or digestibility (Carpenter, 1958). That amino acid composition is not a reliable indicator of protein quality as determined by biological assays may be due to: (1) low digestibility of the protein because of improper processing, (2) reduction in the amount of metabolic activity of one or more critical amino acids, (3) differential rates of hydrolysis and absorption in the gut yielding an amino acid composition unlike that of the native protein, (4) failure to destroy, inactivate, or remove physiologically deleterious substances naturally present in the raw material, (5) contamination with foreign toxic substances during processing or (6) formation of physiologically harmful substances during storage (Duckworth and Woodham, 1961; Bergen, 1971). The nutritive value is also affected by such factors as heat, interactions of com- ponents in the food system, other non-protein components in the diet, and individual physiological variability (Liener, 1958; Kreyl and Barboriak, 1960). Hence, dietary studies are always necessary for a 35 valid evaluation of the nutritive value of a protein source. Correla- tions between the chemical score and various biological assays have been discussed in a FAQ publication (1957) and by Cresta et a1. (1969). The traditional method for determining the biological quality of a protein is to compare the performance of animals (classically rats) on a diet with a recognized good protein such as casein or egg, usually at a level of 10%, with the performance on a diet in which part or all of the "good protein" is replaced by the novel protein (Carpenter, 1951; Pirie, l966e). One of the first studies was that of Davies, Evans, and Parr (1952), who determined that the biological value (BV) and digestibility of whole leaf protein were low. They also noted that the cytoplasmic fraction had a higher BV and digestibility than the chloroplastic fraction and that there appeared to be little difference between plant species. An increase in the age of the plant was accompanied by a decrease in BV and digestibility. That same year Carpenter, Duckworth, and Ellinger (1952) determined, using the gross protein value (GPV) for chicks-~i.e., a comparison of the capacity of the protein under test to that of casein to supplement the cereal protein in rations for growing chicks--that leaf protein fed at low levels (3%) was superior to groundnut meal, but was inferior at higher concentrations (11%). Protein efficiency ratios (PER), true digestibility (TD) and BV determined with rats revealed that leaf pro- tein.was inferior to white fish meal but similar to groundnut meal. Hughes and Eyles (1953a,b) concluded that leaf protein from lucerne was as valuable as fish meal when incorporated as the main supplementary protein in chick rations. However, Cowlishaw et a1. (1954) found lucerne protein concentrate (roller dried at 120 C) to be inferior to fish meal but better than 36 groundnut meal when added as the only supplementary protein to cereal rations for chicks. In a continuation of this study, using the GPV technique of Carpenter, Ellinger, and Shrimpton (1955), in which LPC provides 27% of the total dietary protein, Cowlishaw et al. (1956a) determined that different cr0ps had widely differing nutritive values that in all cases were inferior to casein. Lucerne contained a growth depressing factor (suggested to be saponin) which was counteracted by adding cholesterol. The addition of Lys increased the value of the leaf protein diets indicating low availability of Lys in LPC. A com- plementary study by Cowlishaw et al. (1956b), also using the GPV tech- nique, showed that hot water washing improved the value of lucerne concentrates by removing the water soluble saponins. No appreciable improvements were obtained by varying the pH of precipitation or by solvent extraction. Their results are in agreement with Carpenter, Duckworth, and Ellinger (1954); i.e., in that the GPV of LPC is similar to that of groundnut and less than that of soya bean. The GPV of freeze-dried concentrates were significantly greater than those of roller dried preparations, indicating that the proteins may be damageable by heat. Carpenter and Ellinger (1955) found a highly significant cor- relation between GPV and estimates of available Lys of dietary proteins, thus providing confirming evidence for the postulate that the e-amino group of Lys reacts to form a nutritionally unavailable complex when intact proteins are damaged. Widely varying estimates and low value of the nutritive quality of LPC up to 1957 were attributed to improper processing, especially poor washing and over heating during drying (Buchanan, 1969b). Barber, Braude, and Mitchell (1959) concluded from preliminary tests that wet LPC, when fed as a protein supplement to practical rations 37 (7% LPC) or diets suboptimal in protein (3.5% LPC),was equal to white fish meal for growing pigs. Leaf protein concentrate was tested as a source of supplementary protein in the diets of chicks and growing rats by Duckworth and Woodham (1961). Provided that the method of drying did not involve high temperatures (greater than 81 C), all preparations had uniformly high GPV values (mean of 79) similar to values for soya bean, i.e., 74. In another study Duckworth, Hepburn, and Woodham (1961) found that wheat LPC, when used at a 7% concentration as a source of supplementary protein in the practical diets for newly weaned pigs, promoted the same rate of growth and efficiency of feed utilization as standard diets with 7 or 8% fish meal. The first study utilizing humans as test species was that of Waterlow (1962). Twenty-one Jamaican infants who were recovering from severe malnutrition.were put on diets supplemented with LPC. Since leaf protein might never be used as the sole source of nitrogen in the diet, this study proposed to determine if leaf protein could be used as a supplement to marginal quantities of milk. Therefore, mixtures in which one-half to two—thirds of the nitrogen was derived from leaf pro- tein and the remainder from milk were given with two controls-amilk at an equal nitrogen level and milk at a low level of nitrogen. Balance studies (details given by Waterlow and Wills, 1960, and Waterlow, 1961), which measure the intake and excrement of nutrients, were made. Leaf protein concentrate was tolerated well by the infants and neither weight gains nor nitrogen retention differed significantly between the leaf protein milk (LPM) mixture and milk alone. With low protein milk diets nitrogen retention was nil, but highly significant positive balances were achieved with the addition of leaf protein. Therefore, it was con- cluded that leaf protein could act as a valuable supplement to a marginal 38 intake of milk. The mean values for nitrogen retention are given in Table 7 . Table 7. Mean values for the retention of nitrogen by malnourished infants fed milk (M) or leaf protein and milk (LPM) Group Diet N intake* N retained* Z N retained A1 LPM 727 156 :_21.4 21.4 :_l.8 Bl M 743 175 j; 32.9 23.9 1 3.6 A2 LPM 471 105 :_l6.5 21.2 :_2.5 B2 M 491 108 i 19.0 22.4 i 4.3 Cl M 246 14:20.7 5.4:9.3 C2 M 159 13 :_17.1 6.2 :gl.9 * Expressed as mg N/kg body weight/day In a later study Fox and Waterlow (1967) investigated the value of leaf protein as a partial replacement for dry skim milk (DSM) and as a supplement to the poor Jamaican diet. In animal trials the EV, TD, and net protein utilization (NPU) of the LPC ranged as follows: 41-54, 60-80, and 30-40, respectively; except for banana leaf protein. The suggestion was made that this latter preparation might contain some toxic constituent. Since cow pea LPC appeared to be the most economical to produce locally it was assessed as a supplement. With rats the addi— tion of cow pea LPC to the poor Jamaican diet caused a doubling in per cent weight gain. In long term rat trials no gross post mortem abnor- malities were found.which could be attributed to the cow pea LPC. Balance studies were conducted with normal young children who had recovered from protein malnutrition. The three diets fed were: (1) DSM, (2) l/2 DSM protein and 1/2 solvent extracted LPC protein (LPI), and (3) 1/2 DSM protein and 1/2 LPC protein (LPII). The results are given in Table 8. With the leaf protein diets, in spite of positive 39 Table 8. Mean values for nitrogen absorption and retention by mal- nourished infants fed a milk diet (DSM) and a leaf protein and milk diet (LP) ‘ Evaluation criteria DSM LP I LP II Weight change (g/kg/day) +1.7 -3.0 —0.34 N intake (mg/kg/day) 195 201 196 N absorbed (as Z of intake) 85 :_4.8 87 i 2.1 79 :_5.1 N retained (as Z of intake) 32 i 3.5 33 :_6.9 26 :_4.4 True N digestibility 86.5 86.8 78.8 Net protein utilization 83 83 78 nitrogen retention there was a weight loss. With the exception of weight change there was little difference between DSM and DSM partially substi- tuted with solvent extracted LPC. However, the non-solvent extracted LPC gave the poorest responses. The investigators concluded that it was difficult to make a realistic assessment of the value of cow pea LPC, but it appeared promising if it replaces less than 50Z of the protein of DSM. Henry and Ford (1965) determined BV and TD values for a number of LPC preparations by the method of Mitchell (1924) at an 8Z level of protein intake on young growing rats. In general, BVs of the LPC (over half greater than 70) were of the order found for legumes, cereal seeds and yeast, but the TDs were lower (range of 71-91). Both values varied with species. Freeze-drying, acetone extraction and drying on starch were satisfactory but oven drying at 100 C caused marked decreases in BV and TD. BV and TD increased with maturity of the leaf due to increased availability of Met; when LPC from tares was supplemented with sufficient Met to bring the level equal to that of whole egg, the EV was increased to that of egg. The availability of some essential amino 4O acids was determined microbiologically by the methods of Ford (1960, 1962, 1964) and the results were consistent with the previously stated biological findings (i.e., BV and TD). In agreement with earlier work (Davies et aZ., 1952), both the BV and TD were greater for the cytoplasmic fractions than for the chloroplastic fractions. The results of GPV (modified method of Duckworth, Woodham, and McDonald, 1961) and PER studies by Woodham (1965) confirmed the poten- tially high value of leaf protein. Oke and Tella (1967) showed from feeding experiments with rats that leaf protein was as good as milk protein. Smith (1966) reported that alfalfa leaf protein had a BV, digesti— bility, and NPU of 59, 75, and 44, respectively, for the rat. However, Met supplementation increased these parameters to 72, 77, and 56, respectively. He stated the relatively low digestibility was not associated with processing, drying conditions, or a saponin factor. Nutritive values of leaves and LPC, mostly cassava, prepared in Nigeria were determined by Eggum (1970) using rats. The following ranges were noted: BV, 44-57; TD, 70-78; and NPU, 34-41. The prepara- tion with the highest Met content gave the highest BV. Methionine supplementation of cassava leaves raised its BV from 49 to 80. Cassava leaves combined with Norwegian dried cod, an important protein source in Nigeria, showed a mutual supplementation effect (i.e., the BV of cassava leaves was raised from 49 to 73). Nutritional studies conducted in India generally indicated at least some supplementary value for most leaf protein preparations. Subrahmanyan and Sur (1949) showed that lucerne powder is an excellent supplement to the poor South Indian rice diet when fed at levels of 10% 41 to rats. In 1954 Sur and subrahmanyan reported that lucerne has a sup- plementary value even at low levels. Studies by Shurpalekar, Singh, and Sundaravalli (1967, 1969) revealed a PER (with rats) for freeze-dried lucerne LPC of only 1.54 (skim milk powder 3.3), but supplementation with Met increased this to 2.77. No change in the PER.was noted with Lys supplementation. The greatest growth was found at a 15% protein level when leaf protein was the sole protein source. However, Met sup- plementation gave the best growth at a 10Z protein level. They noted that the addition of cholesterol was unnecessary to counteract the possible presence of growth depressing saponins due to the satisfactory removal of soluble constituents by washing of the curd. Also the quality of leaf protein precipitated by acid was found to be slightly inferior to that coagulated by steam. Leaf protein concentrate--at a 5Z protein level--p1us vitamins and minerals was equal to skim milk powder in its ability to supplement a poor rice diet. Sur (1967) in rat feeding studies with water hyacinth leaf protein found that Met supplementation increased both BV and digestibility. Leaf protein and rice protein had complementary amino acid patterns. Ghosh (1967) also improved the nutritive value of water hyacinth leaf protein by Met supplementation. At CFTRI supplementation with leaf protein led to significant nutrit- ive improvement of rice diets (Singh, 1969). Singh also reported that there were differences in nutritive value between species and between different preparations of lucerne LPC. This latter difference.was attributed to differences in age of the plants. The following PER values were given: carrot, 0; Dhaincha, 0.8; lucerne, l.3-1.8; and crucifers, 1.8-2.2. Histological examination of the liver and intestine of those animals which did not grow did not show any signs of toxicity. 42 Subba Rau at al. (1969), using rats, compared the nutritive values of Spray dried juice, LPC, as well as the chloroplastic and cytoplasmic fractions of leaf protein, prepared from lucerne. Their results are given in Table 9. The values obtained for the whole juice were extremely low. The chloroplastic fraction was inferior to the unfractionated Table 9. Nutritive value of lucerne leaf juice, LPC, and the chloro- plastic and cytoplasmic fractions of leaf protein Protein source PER Apparent digestibility Batch I Spray dried juice 0.34 --- LPC 1.38 :_0.13 --- Batch II . LPC 1.36 :_0.13 73 Chloroplastic fraction 0.78 :_0.13 65 Cytoplasmic fraction 2.02 i_0.13 80 leaf protein which was inferior to the cytoplasmic fraction. The digesti- bility values are in agreement with those reported earlier by Smith (1966) for similar lucerne preparations. The small differences in Lys and Met values between the preparations cannot account for the nutritional variations. The authors suggested that nonprotein constituents act as low digestibility diluents and probably lower the nutritive value through various interactions. Speculation was made that the toxicity of the spray dried juice might be due to high contents of soluble mineral matter and nitrogen constituents, some antigrowth factor, or nutrition- ally unavailable protein damaged by interactions with other substances. This study illustrated the necessity for bioassays; for an in vitro digestion (Hartman et al., 1967) failed to indicate the unsuitability of whole juice as a dietary additive. III ' \II. .I i A! I lif..i.'| It... . ‘ 43 Singh (1967) reported that in a preliminary metabolic study of eight children on LPC diets the TD and BV were 85 and 65, respectively. In a six month trial with children, diets based on ragi--a millet culti- vated widely in South India, contributing a major ingredient in the diet and reported to be deficient in lysine-dwere supplemented with Lys, lucerne LPC, or sesame flour. Nitrogen retention, apparent diges- tibility, height, weight, and hemoglobin were measured. All three supplements resulted in improved nutritional response, but leaf protein gave the greatest growth response because it improved both the quality and content of protein in the diet (Doraiswamy, Singh, and Daniel, 1969; Singh, 1967). However, extensive tests indicating the long term nutritional effects of leaf protein supplementation have not been made (Smith, 1969). In vitro Digestion Methods Various investigators have used in vitro methods for assessing the nutritive value of extracted leaf protein. Akeson and Stahmann (1965) estimated the biological value of freeze-dried LPC with an in vitro enzymatic digestion--based on the amount of eight essential amino acids released by pepsin followed by pancreatin hydrolysis and compared to that released by whole egg. Excellent correlation, better than the essential amino acid index or chemical score, has been obtained between this pepsin pancreatin digest index (PPDI) and published BVs for eight reference proteins (Akeson and Stahmann, 1964). The estimated BVs of LPC (78-89) exhibited little difference between species and in general were less than the values for egg (97), about equal to whole milk (83), and greater than beef (75), casein (76), yeast (71), soybean (65), cottonseed (64) and wheat (50) flours, gluten (45), and zein (26). 1" «I'lllllll. ll'lll‘.[ . 44 This indicated that LPC would be a high quality protein source for human consumption (Stahmann, 1968a,b). Hartman et al. (1967) reported that BV, estimated as PPDI, of spray-dried alfalfa juice samples was similar to those reported for freeze-dried LPC by Akeson and Stahmann (1965). Smith (1966) reported the in vitra digestibility of alfalfa leaf cytoplasmic and chloroplastic fractions to be 95-100 and 60-70, respect- ively. Solvent extraction of the chloroplastic fraction did not increase its digestibility. Byers (1967a) found that maize leaf protein was not digested by papain under conditions in which casein is hydrolyzed (Pirie, 1963). Maximum digestion.was achieved at pH 6.6 and 70 C with KCN-activated papain. Evidence was presented indicating that lipids do not protect the protein against proteolytic digestion. Although in viva experiments have shown higher nutritive value with protein from mature wheat leaves, there was no corresponding increase in the in vitra digestibility. In the following paper Byers (l967b) described the in vitra digestion by papain of LPC extracted from 14 species of plants. The extent of hydrolysis, which was never complete, varied with species. The effect of leaf age was uncertain. For all species the chlorOplastic fraction (precipitated at 53 C) was digested less and the cytoplasmic fraction (precipitated at 80 C) more than the corresponding whole protein. She suggested that the protein digestibility of processed LPC preparations could be assessed rapidly by papain hydrolysis as a preliminary evalua- tion to in viva tests. Pirie (l969e) reported the analyses of seven whole leaf preparations from Mysore by this method. Buchanan (l969b; Pirie, 1967c) compared several in vitra methods with in viva evaluations for estimating the nutritive value of LPC. The thiOglycollic acid-activated papain solubilization procedure of 45 Buchanan and Byers (1969) gave the best correlation with in viva results for rats, but the PPDI seemed promising. Heating moist LPC decreased the TD due to protein-unsaturated lipid complexing, and therefore also the PER and NPU values. The latter two were almost restored to normal values by lipid extraction, since this caused the splitting of these lipid-protein complexes (Buchanan, 1968). Although the sulfur amino acids are usually limiting, Buchanan (l969b) stated that Lys may be limiting if LPC is heated. Shah, Riaz-ud-Din, and Salam (1967) studied the effect of heat on the in vitra digestibility of LPC. Treatment of LPG at 100 C decreased its digestibility with human proteolytic enzymes. Digestibility increased when the LPC was defatted with chloroform-methanol. The oxidation products of the lipids and their polymers with proteins were determined to be toxic to trypsin, pepsin, and pancreatic enzymes. Effects of Storage Pirie (1961c, 1962) reported decreased nutritive value and TD with drying or storage at high temperatures, and suggested that products of lipid oxidation combine with protein breakdown products to form insoluble compounds. Shah (1963) noted that after storage at 80-100 C the extract- ability of LPC lipids decreased. Pirie (l966e) reported that the digesti- bility of barley protein was increased by partially extracting the lipid, indicating that the resistance of leaf protein to proteolysis depends in part on the formation of lipid complexes. This concept was in con- trast to the opinion of Byers (19673). In a study by Buchanan (l969c) at 4 and 28 C, extensive oxidation of LPC occurred, but with little concomitant change in its papain digestibility. Losses in digestibility after heating at 100 C were attributed to two different reactions: (1) 46 a lipid-protein complexing reaction.whidh is reversible by solvent extraction and (2) modification of the protein.which was not influenced by the presence of leaf lipid. It was not possible to conclude how LPC should be stored, only that the interactions occurring during storage are complex. Therefore, Pirie (l966a,c, l969b,d) suggested that variations in nutritive value, despite uniform.amino acid compo- sition, may be the result of changes taking place during preparation or storage due to reaction of carbohydrate, phenolic compounds or fatty acids with amino acid residues. In summary, the several studies showed that leaf protein extracted from many green plants, if not damaged by inapprOpriate processing, has a food value as great as that of fish meal or the best seed protein but less than that of casein and egg (Pirie, l966c, 1967a; Stahmann, 1968a; Kinsella, 1970). Total protein in the diet is rarely from one source, therefore the manner in which the various sources complement each other is as important as the composition of each source separately; e.g., leaf protein is limiting in Met, fish and maize are not (Pirie, 1957b; Marrison and Pirie, 1960; Oke, l966a). Since leaf protein is intended as a supplement to a diet which contains some but insufficient amounts of protein (Pirie, 1959a, 1968), more feeding trials are needed in which leaf protein is compared not only with a reference protein, but combined with other protein that constitute the major portions of local diets. That is, leaf protein should not be considered in isolation, but rather in relation to other dietary components (Pirie, 1968). There is little doubt of the potential value of LPC as a supplement for improving the quality of conventional dietaries based on cereals (Anon., 1970). 47 Indigenous Toxic Substances Many plants have the capacity to synthesize a wide variety of chemical substances which are known to exert a deleterious effect when ingested by man or animals (Liener, 1969). While most plant species undoubtedly contain potentially toxic* compounds, low levels of exposure generally preclude intoxication; for toxicity is determined not only by the intrinsic properties of the toxicant and the exposed individual, but also by the level and duration of exposure (Crosby, 1969). The subject of naturally occurring toxicants in foods has been reviewed by Liener (1962, 1966), Mickelsen and Yang (1966), Strong (1966) and NAS-NRC (1966). This review will concern those which might have possible consequence in the production of protein from green leaves, particularly from alfalfa. Substances which have the ability to inhibit the proteolytic activity of certain enzymes are found throughout the plant kingdom, particularly among the legumes (Liener and Kakade, 1969). Ramirez and Mitchell (1960) described the partial purification of a trypsin inhibitor from alfalfa which they believed to be a non-dialyzable polypeptide or a noncoaguable protein. Mooijman (1964) was of the opinion that the alfalfa trypsin inhibitor is a saponin-peptide or saponin-amino acid complex. Liener and Kakada (1969) discussed the following aspects of protease inhibitors: distribution in plants; physical and chemical prOperties; specificity, stoichiometry, kinetics, and mechanisms of inhibitor-protease reactions; nutritional and physiological significance; and effects of processing. * Toxic is defined as producing an adverse physiological response in man or animal by a particular food (Liener, 1969). 48 Certain grasses, pulses, root crOps, and fruit kernels contain relatively high concentrations of cyanogenetic glucosides (Montgomery, 1969). These plant tissues may be rendered less toxic by extraction or by maceration and dehydration (Conn, 1969). The cyanogens are important in the production of leaf protein from cassava. Their chemistry and biosynthesis have been reviewed by Conn (1969). The family Cycadaaeae which is indigenous to the tropics and possesses large leaves is also toxic when not properly prepared (Mickelsen and Yang, 1968; Yang and Mickelsen, 1969). Natural thioglucosides from the CrucifErae and related plant families are a source of goitrogens or antithyroid compounds (VanEtten, 1969). Included are some common crucifer plants--kale, rape, radish, mustard,turnip-~which have been considered for leaf protein production. Fortunately the thioglucosides occur in highest concentration in the mature seed (VanEtten, 1969). The chemistry and biological effects of thioglucosides (or glucosinolates) have been reviewed by VanEtten (1969) and VanEtten, Daxenbichler, and Wolff (1969). Alkaloids represent a variety of unrelated structural types, common only in the classical sense that they contain nitrogen, have a bitter taste, and are very active biologically, but without showing post- mortem lesions (Sollman, 1957; Keeler, 1969). Alkaloids appear to be the active principals in many range plants which produce acute toxicosis or teratogenic effects (Keeler, 1969). Metabolites detrimental to the growth of meadow vole weanlings were extracted from individual alfalfa clones and were tentatively determined to include the following alkaloids: trigonelline, stachydrine, and homostachydrine (Elliott, 1963b). The toxicity of non-nitrogen containing plant phenols has been reviewed by Singleton and Kratzer (1969). These are toxic only if the 49 natural barriers or detoxification medhanisms are overloaded or evaded. Generally the toxicity limits are not even approached in plant foods, but the potential toxicity should not be neglected (Singleton and Kratzer, 1969). Tannins--polyphenolic substances which have the ability to form a precipitate with gelatin under certain conditions (Kirby, l960)-are present in a number of plant materials at very high levels (Singleton and Kratzer, 1969) and are present in toxic concentrations in certain foods (Click and Joslyn, 1970). Although Kirby (1960) found tannins injected subcutaneously induced tumors in rats, oral administration to mice caused no adverse effects. The chemical and physiological properties of some toxic amino acids and peptides occurring naturally in foods have, along with the mechanisms of toxicity, been reviewed by Hylin (1969) and Fowden, Lewis and Tristam (1967). They point out that most food consumed by man does not contain toxic amino acids and/or peptides. Those which do are usually unique to certain limited climatic areas of the world and are consumed by relatively small papulations (Hylin, 1969). Pudelkiewicz and Matterson (1960) showed that alfalfa contains an ethanol soluble fraction which reduces the availability of vitamin E. Elliott (1963a) found critical levels of a competitive antimetabolite to niacin in certain individual alfalfa plants. Later, a source of diploid alfalfa was found to contain plants with antimetabolic activity as evidenced by bioassays. This activity could be counteracted by a combination of aspartic acid, glutamine, and glycine. Differences in nutritive value between the plants could not be associated with amino acid deficiencies (Schillinger and Elliott, 1966b). 50 Saponins are glycosides that occur in a wide variety of plants. They are characterized by a bitter taste, formation of a stable foam in aqueous solution, hemolysis of red blood cells, formation of molecular compounds with cholesterol and other hydroxy steroids, and an irritating effect on mucous membranes (Sollman, 1957; Hanson et aZ., 1963; Scardavi and Elliott, 1967; Birk, 1969). They exhibit a wide variety of physiological effects, mostly adverse, on animals. Most of these activities arise from the strong surface activity of saponins and from their ability to form complexes with sterols and proteins (Birk, 1969). There are two main classes of saponins: steroids (C27) and tri- terpenoids (C30) (Scardavi and Elliott, 1967; Birk, 1969). They differ markedly in chemical composition with respect to both alglycones and carbohydrate and on hydrolysis they yield sapogenins (the alglycone moiety) plus sugars (Birk, 1969). The chemistry and biosynthesis of the steroidal saponins (and glycoalkaloids) has been discussed by Heftmann (1967). The exact nature of saponins and their potency in foods have been studied relatively little; and some of the properties attributed to them have not always been conclusively demonstrated (Birk, 1969; Gestetner et al., 1970). Among the biological phenomena which may involve alfalfa saponins are growth depressing effects, depressed egg production in laying hens, ruminant bloat, hemolytic effects, reSpira— tory inhibition, nonspecific inhibition of alpha-chymotrypsin, proteases, cholinesterase, and retardation of seed germination (Hanson at al., 1963; Scardavi and Elliott, 1967; Birk, 1969; Shany et al., 1970a). Saponins appear to be partly responsible for the depressing effect of alfalfa on growth, food consumption and diet utilization of chicks. Initial studies included those of Peterson (1950) and Lepkovsky at al. 51 (1950) noted that fractions obtained from an aqueous extract of alfalfa meal produced depression of growth in young chicks proportional to the level of alfalfa meal in the ration. The strong foaming properties of these fractions suggested saponins as the growth depressing agent. All samples having growth depressing action also had hemolytic activity. Most of the inhibitory effect on growth could be removed by eXhaustive extraction of the alfalfa meal with hot water (Lepkovsky et al., 1950). Walter at al. (1954) investigated the water soluble fraction of alfalfa to test the hypothesis that it contains saponin capable of inhibiting growth of chicks and of contributing to ruminant bloat. Coulson and Davies (1962) reviewed the implication of alfalfa saponins in the etiology of bloat and respiratory inhibition. The studies of Coulson and Davies (1962) and Lendahl et a2. (1957) suggested that alfalfa saponins are involved in the formation of bloat in ruminants by altering the surface tension of the ruminant contents. In 1959 Jackson and Shaw compared alfalfa saponins, purified by different methods, on the basis of respiratory inhibiting activity, specific rota- tion, IR spectra, and monosaccharides released on hydrolysis. Pederson et a2. (1966) conducted a comparative study with saponins from different alfalfa varieties of their biological activities. Most saponins are so poorly absorbed in the intestine that acute oral doses produce only local effects on the mucous membranes. Conse- quently, death is due to inflammation of the alimentary canal rather than to absorption and systemic action of the saponin (Sollmann, 1957). Varieties differ in saponin quantity as well as biological value. Hanson et al. (1963) studied the magnitude and variation of the saponin content in alfalfa as a function of location, cutting, variety, and ‘Jl II! {If 52 other variables. The first cutting was lower in saponin content than the last two, which were approximately equal. Saponin content, which varied from 2% to over 3Z, was positively correlated with contents of protein, ash, fat, and nitrogen-free extract; and negatively correlated with crude fiber and hay yield. The chemical composition of alfalfa saponins, which belong to the class of triterpenoids, has been reviewed by Hanson et al. (1963) and Birk (1969). Actually, little work has been done in this area due to the difficulties in purification and characterization and to the hetero- geneity of saponins. By improved methods of paper chromatography, Coulson and Davies (1962) revealed 10 constituents of lucerne bulk saponins. Pederson et al. (1966) found alfalfa to contain a mixture of at least 5 saponins. Methods for their detection, extraction, purification, determination, and Characterization have been reviewed by Scardavi and Elliott (1967) and Birk (1969). But a simple quanti— tative method for the determination of saponins is not yet available (Scardavi and Elliott, 1967). Recently, Shany and associates (Shany at aZ., l970a,b; Gestetner at aZ., 1970) compared the chemical composition and biological activity of saponin preparations from lucerne tops and roots. The root extracts possessed stronger hemolytic activity and surface activity than the top extracts. They found the toxicity of lucerne saponins due mainly to those saponins which contain medicagenic acid or an unidentified sapogenin as their aglycone moiety. It was also reported that the amount of cholesterol needed to fully counteract the saponin-induced growth impairment depends on the toxicity of the extract preparation and on its concentration in the diet. 53 Toxins may also be introduced in a food during processing. Browning, which can be enzymatic or nonenzymatic, may occur due to the formation of unstable polymers of varying composition. It may result in the following adverse effects: (1) development of brown discolora- tion; (2) production of stale, carmelized, or bitter odors or flavors; (3) loss of solubility of the protein; (4) increased tendency to foam; (5) loss of nutritive value of the protein; and (6) possible toxic effects (Friedman and Shibko, 1969). A complete study of the circling syndrome produced in mice by dimethylaminohexose reductone—-a product of the browning reaction--administered orally or by injection, has been reported by Cutting et al. (1960). The syndrome was characterized by head tossing and running in circles. At higher dosages early deaths occurred and at lower dosages the syndrome was absent or incompletely develOped. Ambrose, Robbins, and DeEds (1961) also studied the toxicity of amino-hexose-reductones, administered orally. They noted typical symptoms of hyperexcitability, whirling, and elevations and nodding of the head. Although there is evidence that prOper processing can eliminate toxic substances from LPC, new leaf protein isolates should be examined for toxicants (Buchanan, 1968). Acceptability Evidence of good nutritional value is of little practical signifi- cance if the physical characteristics--color, flavor, and odor--are objectionable to humans. Leaf protein concentrate which has not been solvent extracted has a dark green color and a slight spinach, tea-like or leafy flavor and consequently does not elicit immediate appeal (Byers et aZ., 1957; Morrison and Pirie, 1960; Pirie, 1957b, 1959a, 54 l966a; Buchanan, 1968, l969a; Singh, 1969; Oke, 1969; Kamalanathan, Usha, and Devadas, 1969; Smith, 1969; Knuckles et al., 1970). However, no novel food, regardless of intrinsic merits, is likely to be accepted by adults (Pirie, l969b); for acceptable foods are determined largely by convention. This bias is less pronounced in children (Morrison and Pirie, 1960). Pirie, who for many years has been eating 5-10 g/day quantities of leaf protein (Pirie, l966a,b), maintains that the problem is mainly an aesthetic one, and that it is only a myth that food habits are unchangeable for people have few food instincts (Pirie, 1959a, l966d; Byers at al., 1965; Buchanan, 1968). He does admit that people are reluctant to change and that popularization of novel foods may be difficult (Pirie, 1968, 1969a,d; Buchanan, 1968). Wilson (1964) and Onrust (1969) have discussed general factors which must be considered in attempting a nutritional change with non- conventional proteins. Pirie (l966a,d, 1967b, 1968, 1970) outlined steps for introducing a novel food source: (1) be sure that the novelty can be made in sufficient quantity in the region where it is to be consumed, (2) be sure that it has the feeding value for people that animal experiments suggest, (3) be patient and eat it manifestly, (4) resist attempts to use it first on underprivileged groups, i.e., establish a status associated with the upper class, (5) devise dishes acceptable to those doing and financing the research. After accomplish- ing these steps, acceptance is simply a matter of studying the local food patterns to see which dishes are amenable to modification.‘ Rate and degree of acceptance will depend on dishes normally eaten. Fortunately, tastes in the wet tropics, where the need is the greatest, differ from our own (Buchanan, 1968). Dark colored and greenish foods, as well as powdered dried leaves, are part of the normal diets in 55 India, West Africa, Southeast Asia, and New Guinea--regions where the peOple are primarily vegetarian (Pirie, l966a, l969b; Buchanan, l969a). Guha (1960) stated that color and flavor do not matter when fresh LPC is incorporated into common Indian dishes like curries, pakora, and purries; for they have been well accepted and were, in fact, consumed during the famine in Bengal in 1943. Akinrele (1963) removed the leafy flavor by a thorough water wash and by boiling for a short time or by masking with spices. He and Oke (1966a, 1969) agreed that color should not be a problem in Nigeria where foods of widely varying colors, particularly green vegetable stews and soups, are commonly eaten. However, Akinrele noted that a white LPC, which could be mixed with cassava or yam flour, would be desirable. Oke (l966a, 1968a) included lists of some Nigerian dishes into which freeze-dried LPC was successfully introduced. Singh (1967, 1969; CFTRI, 1968) reported that preliminary trials with several common Indian dishes and some processed foods showed that LPC was acceptable as a base material for incorporation into certain foods, including Chutneys, vegetable curry bases, soup powders, and weaning foods. Mahadeviah and Singh (1968) removed the bitterness of chicory LPC by boiling with water. Kamalanathan at al. (1969) pUblished their results on an investi- gation of the incorporation of LPC into foods, and gave an evaluation of the acceptability of LPC in foods. Leaf protein concentrate prepared from lucerne was incorporated into six preparations at levels of 5 g, 10 g, and 15 g per serving. Appearance and color, texture, flavor, taste, and overall palatability were evaluated. Ratings for the various preparations are given in Table 10. Color was not as great a problem as the leafy flavor. The workers concluded that the flavor of LPC 56 Table 10. Mean hedonic ratings of non-LPC and LPC incorporated products * Preparations Scores for variations Non LPC 5 g level 10 3 level 15 g level Dhal balls . 6.56 5.04 3.80 2.56 'Chutney powder' 6 00 4 88 4.40 4 04 Sweet potato 'curry' 5.76 3 80 3.36 2 68 Ragi 'adai' 6.16 5.48 4.44 4.20 Potato 'bath' 6 28 4 24 3.24 2 64 Greens 'chutney' 6 84 6 00 5.56 4 68 * Maximum score = 7 influenced the preparations adversely and for better acceptability LPC should be incorporated into highly spiced or flavored preparations. In an acceptability trial of SS Jamaican recipes, Fox and Waterlow (1967) found the objection dependent on the recipe in which LPC was used. Color was generally a deterrent. According to Shorland (1969), replacement of flour in sponge cake with alcohol extracted grass leaf protein gave a product which was acceptable except for the detectable residual chlorophyll. In New Guinea, Pirie (l969b) reported that both children and adults ate leaf protein readily. It has also been reported that in New Guinea leaf protein is used for the manufacture of a bread which has been acceptable to the natives (Anon., 1967). According to Parpia (1967) a fair amount of progress has been made in developing acceptable recipes of traditional food preparations incor- porating LPC. Others feel that a better understanding of food habits and traditional methods of food preparation is required for the regions where leaf protein might be used (Pirie, l966c; Buchanan, 1968; Singh, 1969). 57 An investigation of methods for presentation of green LPC-prepared by the method of Morrison and Pirie (l961)-as a food is proceeding at Rothamsted. The first studies were based on the assumption that leaf protein should be used to satisfy 10% of the protein need, about 4-8 g/day (Pirie, 1959c; Morrison and Pirie, 1960). Several recipes--e.g., peanut butter spread, stuffed tomatoes, ravioli, and doughnuts-—have been formulated and published (Morrison and Pirie, 1960; Byers et aZ., 1965). These are categorized into three basic types: (1) dry, stable food, (2) mixtures enclosed in a batter or pastry, (3) soups and stews prepared in the conventional way but with leaf protein added (Byers et al., 1965). Those who see leaf protein regularly accepted it (Pirie, 1959a) and therefore, according to Byers et a2. (1965), the problem is to devise a food sufficiently attractive to appeal to those responsible for the agricultural and food policy of a developing country. There is little information concerning flavors of LPC (Kinsella, 1970). The bitter flavor of forage extracts has been attributed to inorganic ions (especially Zn) whidh are claimed to be removable by electrodialysis and ion-exchange (Jennings, 1949). Buchanan (1968) stated that the flavor differs little between species and may become rancid in unextracted preparations. According to Pirie and associates (Pirie, 1959c; Morrison and Pirie, 1960) flavor is no problem when LPC is encased in pastry. Thorough washing with hot acidified water can even make leaves which have a strong flavor yield a bland or mildly flavored product (Singh, 1969). In general, flavors such as fish, cheese or egg blend better with leaf protein than sweet flavors (Byers et aZ., 1965; Pirie, 1957b). Texture is also important. Fresh leaf protein and properly processed freeze-dried preparations can be con- verted to a smooth paste (Byers et aZ., 1965). 58 There appears to be slight variation in the color of LPC among species of plants (Buchanan, 1968). The dry, gray-brown powder made by solvent extraction is easier to incorporate into foods than the green concentrate (Byers at al., 1965). However, the following reasons were given for not using it: (1) it involves sophisticated technology, whereas the objective is to devise methods of producing and utilizing a protein supplement in technically unadvanced countries; and (2) the color has not been found to be a serious obstacle. Singh (1967) stated that the predominant feeling among the experts that only white bland products would be acceptable as protein supplements must be overcome. However, Fox and Waterlow (1967) in Jamaica and Smith (1966) in England agreed that although it may be possible to conceal small amounts of the green LPC in certain highly flavored dishes, for most human feeding purposes the strongly flavored components must first be removed. Opinions in the United States also generally support the deve10pment of a bland, light colored concentrate which can be incorporated into com- mon foods without vastly changing their physical properties (Kinsella, 1970). Potential The potential of leaf protein as a human food has been discussed by many writers. Slade (1937) stated that if used as "grass cheese" four to five times more food per acre could be produced than produced by raising wheat. The premise that it would be more efficient to con- vert the protein in leaf into a human food by industrial, biochemical methods rather than by the ruminant has been expounded many times (Slade at al., 1945; Pirie, 1942a, 1953, 1956b, 1957a, 1959a, 1967a, 1968; Akinrele, 1963; Tilley at al., 1954; Raymond and Tilley, 1956; West, 59 1968; Stahmann, l968a,b; Kinsella, 1970). Biological conversion is very wasteful. The conversion efficiency of the ruminant is never greater than 30% and is usually 5-15Z. In contrast to Pirie's (l942a,b, 1953) early optimistic advocation of leaf protein as a human food, Raymond and Tilley (1956) stated that few papers contained sufficient data to substantiate their claims to the promising future of leaf protein; and that there was no reason to recommend it over legumes or oilseeds. However, under conditions of inadequate food supplies, any method capable of making more food should be of interest. They (Tilley and Raymond, 1957) concluded that the utilization of LPG as a major agricultural product is dependent on the demonstration that the protein could be produced cheaper and in greater yield than alternative sources of protein of equivalent nutritive value, and that it could be converted into a satisfactory food product. Akinrele (1963) felt that its prohibitive cost and restricted availability relative to other protein sources will keep the production of leaf protein a problem of academic rather than practical interest. However, he added that it is a potentially important protein reserve which should not be forgotten. An opposing view is held by another Nigerian investigator, Oke (l968a,b), who feels the main hope for needed protein supplementation lies in the protein of green leaves. He sug- gested that if the three main craps of Nigeria--cassava, maize, and cow pea--are planted for seeds, the yield of crude protein per acre would be about 8, 40, and 100 kg, respectively. Whereas, if they are harvested for their leaves the corresponding yields will be about 70, 200, and 400 kg, reSpectively. In 1964 it was the opinion of the FAO that although there were a nunber of problems-~unattractive color, strong hay taste, and financial 6O support-~to be solved before a useable, low cost product could be pro- duced, the potential food value of LPG is unquestionable. Pirie (l966a) stated that the concept of leaf protein signifi- cantly alleviating the world protein shortage has not yet gained suf- ficient acceptance to initiate large scale production. He feels that emphasis should be placed on the fact that palatable leaf protein, with sufficient nutritive value, can be made in bulk from a number of plants. Inertia and a plentiful supply of protein in those nations capable of carrying on the research have thus far reduced investigative efforts. Akeson and Stahmann (1966) concluded that further study of leaf pro- tein production should be undertaken in order to increase the protein supply in those areas in which it is inadequate. However, Parpia (1967) felt that the utilization of LPC on a large scale is beset with several problems such as keeping quality, taste, and acceptability. According to Singh (1967), the results of the studies on LPC com- pleted thus far have given substantial support to the expected poten- tialities inherent in the direct exploitation of green vegetation. He concluded that a program utilizing some legume fodder plants and the by-product vegetation of several green vegetable crops would be workable for leaf protein production without changing current Indian agricultural practices. Gonzales at al. (1968) in the Philippines stated that leaf protein can be considered a valuable supplementary food in their country. The International Symposium of Protein Foods and Concentrates (1967), feeling that LPC has potential that merited further investigation and study, went on record to support research on leaf protein. The Presi- dent's Science Advisory Council (1967) reported that LPC should be given consideration primarily because the raw materials are frequently wasted or are at best fed to animals with concomitant inefficient utilization of the protein. 61 Stahmann (1968b) speculated that enough alfalfa leaf protein to meet world needs could be produced on only 302,000 square miles. An equal amount of protein would remain unextracted in the residue. This could be fed to cattle to produce additional meat and milk. Although only speculation, it does demonstrate the vast potential of forage pro~ tein if utilized as a primary food source (Kinsella, 1970). Browning (1970) declared alfalfa and grasses to be the most promising sources for economical production of edible LPC. Stahmann (1968b) concluded that alfalfa has the greatest potential of the common crOps for produc- ing a maximum quantity of high quality protein. He is convinced that, in the future, mechanical leaf protein production will be utilized and will assume a significant role in meeting the increasing protein require- ment. It is the Opinion of Smith (1969) that mechanical extraction of protein from leaves promises to provide a cheap source of concentrated protein that will rival traditional animal sources. He felt that North Americans must be persuaded to adOpt leaf protein additives, thus setting an example for developing countries. Shorland (1969) and Kinsella (1970) also concluded that plant leaves can supply an important quantity of protein for future requirements. Although LPC has unquestionably been shown to have potential as a protein supplement, whether or not it will realize this potential will depend to a large extent on whether or not it can meet the criteria by which a food protein is judged--nutritional value, functionality as an ingredient, compatibility with other foods and food ingredients, keeping quality, sensory properties, and social acceptability (Kinsella, 1970). Considering the potential of leaf protein, especially in trapical areas, research efforts are still extremely small (Pirie, l969a; Floyd, 1970). This contrasts to the much larger efforts being given to other potential 62 protein sources: oilseeds, etc. (Singh, 1967; Pirie, 1967a, 1969d; Buchanan, 1968). This disparity should be recognized when quality com- parisons are made of different end products. Pleas have been made for international, cooperative research and training of scientists in the methods develOped for leaf protein production (Pirie, 1967a; Kinsella, 1970). But research is still needed, particularly in crap husbandry, preservation, biological assessments of nutritive value, economics, and acceptability and presentation, as well as on many unexamined species of plants (Buchanan, 1968; Stahmann, 1968b; Singh, 1969; Kinsella, 1970). However, the nations most in need of the results of this research lack the resources to conduct it (Stahmann, 1968a). EXPERIMENTAL Chemicals and Materials Chemicals The principal chemicals used in this study and their sources are listed in the Appendix, Table A1. All were reagent grade unless other- wise stated. Water was distilled and deionized. For the enzyme extractions the following enzyme preparations were specially procured: (l) Wallerstein Cellzyme (R) from Wallerstein Co., Morton Grove, 111.; (2) cellulase, purified (15,000 units/mg), T. viride, from Seikagaku Fine Biochemicals, Seikagaku Kogyo Co., Tokoyo, Japan; (3) alpha-amylase (liquifying), 3x crystallized (643 units/mg), B. subtilis from Seikagaku Fine Biochemicals; (4) pectinase, purified (1.1 unit/mg), fungal origin from Sigma Chemical Co., St. Louis, M0. The cellulase and alpha-amylase were purchased from Miles Laboratory, Elkhart, Indiana. Alfalfa There were three major sources of the fresh green alfalfa. The first was from a green house during winter and spring, 1970. The second and third were Fields A and B available during the summer season, 1970. All three areas were under the care of the Department of Crop and Soil Sciences at Michigan State University. This alfalfa included Madiaaga gZutinasa and primarily Mediaaga sativa, varieties Vernal, Culver/Vernal, MSB, MSA, and Hardy moapa. All had been genetically selected to be low or medium low in saponin level. 63 64 "Greenhouse" alfalfa was used for initial investigations of: the standard extraction procedure, yields, and enzyme extractions (Modifi— cations A, B, C, D). Field A provided the raw material for further yield studies and for extraction of the two fractions which were used for chemical analyses and biological assay. The first-cutting alfalfa from Field B was the source for further enzyme extractions (Modifica- tions E, F, G). The fractions for organoleptic evaluation were obtained from the second cutting. The last cutting of this field furnished alfalfa for extractions by the method of Pirie and Modification I. Preparative Procedures Standard Extraction Method, Modifications A, B, C, D The standard extraction method, Modifications A, B, C, and D, are illustrated in Figures 5, 6, 7, 8, and 9, respectively. The figure descriptions are intended to be a part of and to be read in sequence with the remainder of the text. 65 Figure 5. Flow diagram for standard extraction method. Freshly cut alfalfa was blended with water in a Waring, commercial- sized blender, model CB-3 for 15 sec on low speed, 15 sec on medium speed, and 20 sec on high Speed. Blending was most satisfactorily achieved with units of 200 g of alfalfa plus 800 m1 of water. Two "units" of leaf pulp were poured into a press cylinder which was placed in a plastic pail with an adjustable outlet at the bottom to aid in the collection of the juice. Pressure was applied to the constrained pulp with a Carver laboratory press, model B. The juice was heated with a magnetic stirrer-hot plate. The time required to induce chlorOplast coagulation varied with the quantity of the juice, but usually was about 10-15 min for 3500 m1 of juice. Cen- trifugation was principally performed in an International model V, size 2, with rotor #226, and occasionally performed in a Sorvall RCZ-B refrigerated centrifuge equipped with a type GSA rotor. Adjustments in pH of supernatants l and 2 were monitored with either a Beckman Zeromatic pH meter or a Corning Model 12 pH meter. The two final fractions of interest were precipitate 3 (P3 or the acid precipitate) and supernatant 3 (S3 or the acid supernatant). The latter fraction was condensed in a laboratory constructed vacuum con- densing unit operated at a vapor temperature of about 23 C. The rate of evaporation was 0.7 to 1.0 liter/hr. The fraction was concentrated about 10:1 and dialyzed, with dialysis water changed about every 6 hr. Three freeze-driers were used for lyophilization of the alfalfa fractions: (1) a laboratory constructed lyophilizer which was connected to a Cenco Hyvac 28 vacuum pump; (2) a Virtis Unitrap with combination bulk and manifold freeze-drying chamber connected to a DuoSeal vacuum pump, model 1397 by Welch Scientific Co.; (3) a Virtis RePP #FFD 42 WS freeze-dryer, with a capacity of 50 lb of water removal per drying run and equipped with instrumentation for control of freeze-drying variables. 66 FRESH ALFALFA Cut into lengths of approxi- mately 2 in Blend 50 sec with 4x its weight of water Press pulp at 10,000 psi for 2~3 min JUICE ‘ FIBER RESIDUE Heat to 48 C Discard Centrifuge at 750 g 15 min SUPERNATANT 1 ($1) PRECIPITATE 1 (Pl) (chloroplastic) pH to 8.5 with 20Z NaOH Centrifuge at 750 g 15 min LyOphilize or discard SUPERNATANT 2 (52) PRECIPITATE 2 (P2) (alkali) pH to 4.5 with conc HAc Centrifuge at 750 g 15 min Lyophilize or discard SUPERNATANT 3 (S3) PRECIPITATE 3 (P3) (acid) (acid) Condense under vacuum to Lyophilize about 1 liter Dialyze against distilled water for 48 hr LyOphilize Figure 5 67 Figure 6. Flow diagram for Modification A of the standard extrac- tion method. Fresh alfalfa was cut, blended, and pressed as in the standard extraction method. Fifty milliliters of unheated juice were allowed to remain at room temperature for nonprotein nitrogen (NPN) determina- tions after 0, l, 3, and 8 hr. The bulk of the juice was heated as previously described. Fifty milliliters were held at room temperature for NPN determinations as above. Precipitates l and 2 were separated from the remaining juice by the standard extraction method. Supernatant 2 was split in half (32 and 82'). Fraction 82 was acidified and centri- fuged, yielding precipitate 3A and supernatant 3A (83A). Fraction 83A was heated in a water bath and centrifuged, giving precipitate 3B and supernatant 3B (83B). The precipitation process was then reversed with fraction 82', which was first heated. The resulting supernatant 3A' was acidified. The final supernatants (i.e., 83B and S3B') were passed over a mixed bed ion exchange resin prior to concentration and lyophilization. A yield study was conducted simultaneously (see Figure 10 for flow diagram and for total solids (TS) and nitrogen (N) sample sizes). -—_————.’-—‘_.— 68 ALFALFA (cut) TS, N Weight (blend, press) I 4] JUICE FIBER RESIDUE Volume Weight TS, N (discard) -—-) 50 ml, NPN (5 ml) re- moved at 0, l, 3, 8 hr (heat) —-)50 ml, NPN (5 ml) re- moved at 0, l, 3, 8 hr (centrifuge) I V—l SUPERNATANT l PRECIPITATE 1 Volume Weight TS, N TS, N (alkali, centrifuge) (1y0philize) F 7 SUPERNATANT 2 PRECIPITATE 2 Volume Weight TS, N TS, N (lyophilize) SUPERNATANT 2 pH to 4.2 with conc SUPERNATANT 2 ' Heat 30 min in 95 C water HAc bath (centrifuge) (centrifuge) _1 F I—l SUPERNATANT 3A PRECIPITATE 3A SUPERNATANT 3A' PRECIPITATE 3A' Volume Weight Volume Weight TS, N TS, N TS, N TS, N Heat 30 min in (lyophilize) pH to 4.2 with (lyophilize) 95 C water conc HAc bath (centrifuge) (centrifuge) r** l r 4‘1 SUPERNATANT 3B PRECIPITATE 3B SUPERNATANT 313' PRECIPITATE 33' Volume Weight Volume Weight TS, N TS, N TS, N TS, N pass over ion (lyophilize) pass over ion (lyophilize) exchange bed exchange bed FILTRATE FILTRATE Volume Volume TS, N TS, N (condense, lyophilize) Figure 6 (condense, lyophilize) 69 ALFALFA (cut) TS, N Weight (blend, press) JUICE FIBER RESIDUE Volume Weight TS, N ——> 150 ml, NPN (5 ml) removed at 0, 0.5, l, 2, 3, 4, 6, 8, 24 hr (heat) ——> 150 ml, NPN (5 ml) removed at o, 0.5, 1, 2, 3, 4, 6, 8, 24 hr Remainder as in Figure 10. Figure 7. Flow diagram for Modification B of the standard extraction method. A simultaneous yield study (see Figure 10) was conducted with this modification as with Madification A. Fresh cut alfalfa was blended and pressed by the standard extraction method. One hundred fifty milliliters of the unheated juice were set aside at room temperature from which samples were removed for NPN at the following times: 0, 0.5, 1, 2, 3, 4, 6, 8, and 24 hr. The remaining juice was heated as in the standard procedure and a 150 ml portion set aside for determina- tions of NPN at the same time intervals as the unheated juice. The remainder of the extraction was as in a standard yield study. 70 Figure 8. Flow diagram for Modification C of the standard extraction method. Fresh cut alfalfa was sampled for nitrogen and TS and weighed into two 300 g units. To each unit was added 1200 ml water and 7.5 ml Cellzyme (R). An additional unit of 165 g alfalfa plus 660 ml water was processed without added enzyme and served as a control. Each was blended and pressed as in the standard extraction procedure. Juice from the two enzyme units was combined. Samples for TS and N analyses were taken from each experimental batch (i.e., enzyme and control). The enzyme-treated juice was divided into four approximately equal volumes which were held at room temperature for 0, 4, 6, or 8 hr. The control was split in halves which were held for 0 or 8 hr under the same con- ditions as the enzyme-treated juice. At the conclusion of the holding periods, each juice sample was heated and centrifuged as in the standard extraction method. Samples were taken from supernatant l for TS, N, and NPN determinations and from precipitate l for TS and N analyses. Then the standard extraction procedure was followed, yielding super- natant 3 and precipitate 3. Samples from these were taken for both TS and N, as well as a sample from the supernatant for NPN. 71 ALFALFA (cut) N—250 mg, TS-5 g Blend: 2 units (300 g + 1200 ml water + 7.5 m1 Cellzyme) 1 unit (165 g + 660 ml water) (press) JUICE FIBER RESIDUE Enzyme: N—2 ml, TS—25 ml (discard) hold for 0, 4, 6, 8 hr Control: N-2 ml, TS-25 ml hold for 0 or 8 hr (heat, centrifuge) [ 1 SUPERNATANT l PRECIPITATE 1 TS-25 ml, N-2 ml, NPN-5 ml TS-S g, N-500 mg (alkali, centrifuge) SUPERNATANT 2 PRECIPITATE 2 (acid, centrifuge) discard SUPERNATANT 3 PRECIPITATE 3 123-25 ml, N-4 ml, NPN-5 ml TS-5 g, N-100 mg Figure 8 72 Figure 9. Flow diagram for Modification D of the standard extraction method. Fresh alfalfa, prepared for extraction and sampled for TS and N, was blended and pressed as in the standard extraction procedure. Samples from the juice were taken for T8 and N analyses and six 450 ml volumes of juice were measured. Two of these served as controls and were held at room temperature for O or 8 hr. Two and one tenth milligrams of cellulase (0.465 mg/100 ml juice) were added to each of the remaining four portions which were held at room temperature for 0, 4, 6, or 8 hr. At the conclusion of each holding period, the juice was heated and centrifuged as for the standard extraction method. Samples for TS and N determinations were taken from supernatants l and precipitates 1 obtained from the juices held the times indicated in Figure 9. The standard extraction method was followed to produce supernatants 3 and precipitates 3. Samples for TS and N determina- tions were also removed from the latter two fractions obtained from juices held the times indicated. 73 ALFALFA (cut) N-250 mg, TS-S g (blend, press) JUICE , FIBER RESIDUE TS-25 ml, N-2 ml (discard) Control: hold 0, 8 hr Enzyme: hold 0, 4, 6, 8 hr (heat, centrifuge) [ SUPERNATANT l PRECIPITATE l TS-25 ml, N-2 ml TS-S g, N-SOO mg at t = 0, 4, 6, 8 hr at t - 0, 8 hr (alkali, centrifuge) I SUPERNATANT 2 PRECIPITATE 2 (acid, centrifuge) Discard I SUPERNATANT 3 PRECIPITATE 3. TS-25 ml, N-4 ml TS-S g, N-100 mg at t = 0, 8 hr at 0, 4, 6, 8 hr Figure 9 74 Modification E Modification D was followed with one exception--l mg cellulase was added per 100 ml juice. Modification F Modification D was followed with 10 mg alpha-amylase instead of cellulase added per 100 ml juice. Modification G Modification D was followed with 10 mg pectinase in place of cellulase added per 100 m1 juice. Modification H Two freeze-dried samples (from the standard extraction procedure) of acid precipitate (0.2 g) and acid supernatant (0.1 g) were weighed. To one of each fraction was added 10 ml of isopropanol. Ten milliliters of acetone-haxane (1:1 v/v) were added to the other. The samples were mechanically shaken for one hour and then centrifuged 5 min in a clini- cal centrifuge. The supernatant was poured off and the precipitate dried under vacuum. Modification I Juice was obtained from fresh alfalfa by the standard extraction procedure. To a series of test tubes (8) containing 10 ml of fresh raw skim milk were added in duplicate either 0, 0.5, 1.0, or 2.0 m1 of juice. One set was allowed to stand at room temperature and the other in a 37 C water bath, both for 6.5 hr. The two samples which showed coagulation by the end of this time period (milk + l or 2 m1 juice, incubated at 37 C) plus two controls-- (1) raw skim milk and (2) alfalfa juice-~were centrifuged at conditions 75 to yield a 60 S pellet (Beckman-Spinco L2-65 preparative ultracentrifuge with a No. 65 fixed angle rotor with cellulose acetate tubes, 202,000 g for 76 min). High voltage paper electrOphoresis assays were made on the supernatants . Yield Study and Pirie Extraction The procedures followed for yield studies and the extraction by modification of the method of Morrison and Pirie (1961) are illustrated in Figures 10 and 11, respectively. Again the figure captions are to be read as text. 76 Figure 10. Flow diagram for Yield Study of standard extraction method. The standard extraction method (or one of its modifications) was followed. Total weight of the fresh alfalfa and fiber residue and total volume of the juice were determined. Then 1500 m1 of the juice were used as a starting volume for the remainder of the yield study. As illustrated in the flow diagram, each solid fraction was weighed and the volume of each liquid fraction measured. Representative samples were then removed for T8 and N determinations. Sample sizes are given in this figure. The volumes of alkali and acid necessary for pH adjust- ment were noted. 77 ALFALFA (cut) TS-3 g, N-250 mg Weigh (blend, press) JUICE Volume TS-25 ml, N—2 ml (heat) Centrifuge 1500 ml in tared 250 ml centrifuge bottles I FIBER RESIDUE Weigh (discard) SUPERNATANT l Volume TS—25 ml, N-2 ml (alkali) Centrifuge in tared 250 ml centrifuge bottles 7 PRECIPITATE 1 Weigh by difference TS-5 g, N-500 mg I SUPERNATANT 2 Volume TS-25 m1, N-2 ml (acid) Centrifuge in tared 250 ml centrifuge bottles 1 PRECIPITATE 2 Weigh by difference TS-S g, N-900 mg SUPERNATANT 3 Volume TS-25 m1, N-4 ml (condense, dialyze, and lyophilize) Figure 10 PRECIPITATE 3 Weigh by difference TS-5 g, N-100 mg (lyophilize) 78 ALFALFA Cut Blend with water Press JUICE ‘ FIBER RESIDUE pH to 3.75 with conc HAc Discard Centrifuge at 750 g for 15 min SUPERNATANT PRECIPITATE Discard Lyophilize or Solvent extract and lyophilize Figure 11. Extraction by modification of method of Morrison and Pirie (1961). The method was essentially that described by Morrison and Pirie (1961). Fresh alfalfa was cut, blended with water, and pressed as in the standard extraction procedure. The juice was acidified, the suspen— sion centrifuged, and the supernatant discarded. One third of the precipitate was lyophilized directly. The remaining portion was extracted with isopropanol or with ethanol followed by hexane prior to lyophilization. To one half (about 5 g) 100 ml of isopropanol were added. The alcohol-preparation mixture was mechanically shaken for 1 hr, centrifuged at 750 g for 20 min, and the supernatant discarded. This process was repeated three times. To the other half 100 ml of ethanol were added. The mixture was shaken, centrifuged, and the solvent decanted once as described for the iSOpropanol extraction. Then, a 100 m1 portion of hexane was added to the precipitate which was again shaken 1 hr, centrifuged, and the supernatant decanted. Finally, the two solvent extracted precipitates were lyophilized. 79 Field Harvests Three cuttings (June, July, October) were made of the 1970 Field A of individually planted low-saponin alfalfa (refer to materials section, p. 63, for detail). The first cutting consisted of six harvests (June 5, 6, 7, 9, 11, and 12) of 5 to 6 1b alfalfa each. The second cutting also included six harvests (July 17, l8, 19, 21, 22, and 24) of approxi- mately the same size. Both cuttings were made at the one tenth bloom stage. Light frosts and the potential of a killing frost required the third cutting to be made at late bud—early flower stage (October 2 and 3). The standard extraction method was employed for all collections. Precipitates l and 2 were discarded. At the end of each cutting the freeze-dried acid precipitates of all harvests were combined. Similarly, the freeze-dried acid supernatants of all harvests of a cutting were combined. Six fractions were thus obtained: 1P3 -- acid precipitate of the first cutting 183 -- acid supernatant of the first cutting 2P3 -- acid precipitate of the second cutting 283 -- acid supernatant of the second cutting 3P3 —- acid precipitate of the third cutting 3S3 - acid supernatant of the third cutting These were used for chemical analyses and biological assays. Concomitant yield studies were conducted with the first and fifth harvests of the first cutting, the second and fifth harvests of the second cutting, and the second harvest of the third cutting. Eight harvests (about 5.5 to 6 1b alfalfa each) were made of the second cutting of Field B: August 27, 28, 29, 31, September 1, 3, 4, and 5. The one-tenth bloom stage again indicated the time of harvest. The standard extraction procedure was followed. As in the previous field harvests precipitates l and 2 were discarded. The acid precipitates from the eight harvests were combined; as were the acid supernatants. 80 The latter two fractions served as the sources of supplementary protein fractions for the organoleptic evaluation. Chemical Methods Nitrogen Nitrogen was determined by the micro-Kjeldahl method. To each micro-Kjeldahl digestion flask were added about 15 mg of protein and 4 ml of digestion mixture. (The digestion mixture consisted of 5.0 g of CuSO4'5 H20 and 5.0 g SeO2 The digestion was carried out over a gas flame for 1 hr. After cooling, in 500 ml of concentrated sulfuric acid.) 1 ml of 30Z hydrogen peroxide was added to each flask and the gas heated digestion was continued for another hour. Each flask was then cooled and rinsed with 10 ml water, after which the digestion f1ask.was clamped directly on the micro-Kjeldahl distillation apparatus. The digested sample was then neutralized with 25 m1 of 40Z (w/w) sodium hydroxide solution. The released ammonia was steam distilled into 15 m1 of 4Z (w/v) boric acid containing 5 drops of mixed indicator (400 mg bromocresol green and 40 mg methyl red WS in 100 m1 of 95% ethanol). The indicator is blue in alkaline media, green at the end point, and yellow in acid media. The distillation was continued until a final volume of 70 ml was collected in the receiving flask. The ammonium- borate was titrated with 0.02 N hydrochloric acid either manually or with a Fisher Automatic Titrimeter, Model 36. The hydrochloric acid was previously standardized with tris (hydroxymethyl) aminomethane using p—sulfa—o-methoxybenzeneazodimethyl-l-napthylamine. A reagent blank along with a tryptophan standard were analyzed to determine the average per cent recovery of nitrogen. 81 Nonprotein Nitrogen Five ml of a 30Z (w/v) trichloroacetic acid (TCA) solution were added to five ml of the sample. The mixture was shaken and allowed to stand at room temperature for 30 min. The diluted sample was centri- fuged 5 min in an International clinical centrifuge, model CL. Five milliliter aliquots of the resulting supernatant were used for nitrogen determination by the previously described micro-Kjeldahl method. Amino Acids Amino acid analyses were performed on 22 and 70 hr acid hydrolysates of fractions from the first cutting, but on only 22 hr acid hydrolysates of fractions from the second and third cuttings. The amino acids were separated by column chromatography and quantitatively determined by automatically recording the intensity of the color produced by their reaction with ninhydrin (Moore and Stein, 1954, 1963; Moore, Spackman, and Stein, 1958; Spackman, Stein, and Moore, 1958) with a Beckman 120C Amino Acid Analyzer. Samples containing 15 mg of protein were placed into 10 ml glass ampules. To each ampule 5 ml of once-distilled 6 N hydrochloric acid were added. Prior to deoxygenation the ampules were placed in a Sonogen ultrasonic cleaner by Branson Ultrasonic Corp. for 1-2 hr to break up the insoluble sample residues. Then the ampule contents were frozen in a dry ice-ethanol bath. Using a high vacuum pump, the ampule was carefully evacuated and warmed until all dissolved gases were removed from the liquid sample-acid mixture. It was necessary to add some anti- foam to the neck of the ampule to prevent excessive foaming during this step. The ampule contents were then refrozen and the ampule sealed over an air-pr0pane flame. The sealed ampules were placed in an oil bath [all]. Ill. I I 82 set in a 110 i 0.1 C hydrolysis oven--Stabil-Therm Gravity Oven con- structed by the Blue M Electric Co.—-for 22 or 70 hr. The cooled ampule taps were broken open and 1 ml of norleucine standard (2.5 umole/ml) was added to each as an internal standard to measure transfer losses. The hydrochloric acid was removed from the hydrolysate in a pear shaped 25 ml flask, connected to a Rinco rotary evaporator under vacuum, and partly immersed in a 50-55 C water bath. Antifoam in the stem of the connector prevented the sample from foaming out of the flask. The residue was washed with about 10 ml water and reevaporated. This was done three to four times in succession until hydrochloric acid could no longer be detected. The acid free hydrolysate residue of each sample was dissolved with 3-4 ml of diluter buffer (0.067 M sodium citrate-HCl, pH 2.2). Each sample solution was quantitatively trans- ferred to a 5 m1 volumetric flask and diluted to volume with buffer. To rid the samples of interfering humin they were centrifuged 10-15 min in an International clinical centrifuge. One tenth milliliter aliquots of the supernatant were applied to the analyzer columns. The resulting chromatograms were compared to those of standard amino acid calibration mixtures. The ratio of areas under the curve of each amino acid for the samples and the standard were compared and converted to give the amino acid composition of the sample. For those samples hydrolyzed both 22 and 70 hr, corrections for losses of threonine, serine, and tyrosine during acid hydrolysis were made using the equa- tion given by Hirs, Stein, and Moore (1954). The other amino acids were determined as the simple average of the 22 and 70 hr results. All calculations were done by a computer program written in this labora- tory for use in the Michigan State University computer laboratory. 83 Methionine and Cystine Methionine and cystine undergo variable destruction during acid hydrolysis. Therefore, they were oxidized to methionine sulfone and cysteic acid by the procedure of Lewis (1966) using the performic acid reagent of Schram, Moore, and Bigwood (1954). Approximately 60—65 mg samples of protein were weighed into small sample bottles. These were placed in an ice bath and cooled to 0 C. Ten milliliters of 30Z hydrogen peroxide and 90 ml of 88% formic acid were mixed and allowed to stand at room temperature for 1 hr. The performic acid was then cooled to 0 C in the ice bath. Ten milliliters of performic acid were then added to each sample. The sample bottles were kept in an insulated crushed ice bath inside a 4 C cold room for 17 hr. Twenty milliliters of ice cold water were added to each sample. An unsuccessful attempt was made to freeze-dry the samples with a trap set up to prevent the performic acid from entering the lyophilizer. Each was then diluted with an additional 180 m1 of water. The samples were split so that there were only 50 ml of the oxidized protein suspension per 250 ml bottle. Thus, the samples could be frozen in a very thin shell. However, the samples still did not remain frozen after being placed under vacuum. As a workable procedure, samples were placed immediately on the lyophilizer without the intermediate trap, realizing that in the future a trap would have to be built between the lyophilizer and the high vacuum pump. Frequent oil changes and a cleaning of the pump after operation were necessary. The freeze-dried protein samples were then analyzed by the procedure described for amino acid analyses. Two tenths milliliter of the final supernatants was applied to the analyzer column. Standards 84 containing cysteic acid, methionine sulfone, and methionine sulfoxide were analyzed. Normalized values--corrected for sample weight, nor- leucine recovery, volume on the column, and incomplete oxidation-for cysteic acid and methionine sulfone of the oxidized samples were substi— tuted for cystine and methionine, respectively, of the original unoxidized samples. Tryptophan This amino acid is destroyed by acid hydrolysis. Procedure W of Spies (1967) was used for its chemical determination separately. One to five milligram samples were weighed into 2 ml glass vials with screw caps. To each 0.1 ml of pronase solution was added. (The pronase solu- tion was made by adding 10 ml of 0.1 M sodium phosphate buffer, pH 7.5, to 100 mg pronase. The suspension was shaken gently for 15 min and clarified by centrifugation for 5 min in an International clinical centrifuge. The pronase solution was prepared fresh every day that it was used.) The closed vials were incubated 24 hr in a 40 :_l C water bath. They were then quickly cooled to room temperature in a crushed ice bath. To each vial 0.9 ml of 0.1 M sodium phosphate buffer, pH 7.5, was added. In 50 ml Erlenmeyer flasks 30 mg of p-dimethylamino- benzaldehyde and 9.0 ml of 21.2 N sulfuric acid were mixed. The Open vials were set in the Erlenmeyer flasks, tipped over, and mixed quickly by gentle swirling. The flasks were stappered and placed in the dark at 25 C for 6 hr. Then, 0.1 m1 0.045Z (w/v) sodium nitrite was added to each flask. After 30 min, the per cent transmittance through 1 cm silica cuvettes was determined at 590 nm on a Beckman DK-2A ratio recording spectrophotometer. Simultaneously, duplicate samples of the pronase solution, without protein, were treated and analyzed as above. 85 The tryptOphan content of the pronase solution was subtracted from the total tryptophan value of the protein sample. A standard curve covering the range of 0-200 ug of tryptophan was made according to Procedure E of Spies and Chambers (1948). Four milli— grams of tryptophan were weighed into a 200 ml volumetric flask which was brought to volume with 19 N sulfuric acid containing 3 mg p-dimethyl- aminobenzaldehyde per m1. Zero, 1, 2, 4, 6, 8, and 10 ml aliquots were added to 25 ml Erlenmeyer flasks. The total volume was brought to 10 ml with the same 19 N sulfuric acid. The stoppered flasks were placed in the dark at 25 C for 6 hr. Then 0.1 ml 0.045Z (w/v) sodium nitrite was added and the per cent transmittance at 590 nm read after 30 min. Carbohydrate Carbohydrate was determined by the method of Dubois at al. (1956). One half to two milligrams of the acid precipitate or 0.07 to 0.45 mg (3.5 mg/25 ml) of the acid supernatant were added to a 2.5 cm x 19.5 cm test tube with 3 ad water. Three milliliters of 5Z (w/v) phenol (in water) solution were added and mixed with the sample by hand shaking. Fifteen milliliters of concentrated sulfuric acid were added as the tube was rotated by hand to give maximum mixing. The tubes were allowed to stand for 10 min at room temperature; then, after mixing again, the tubes were placed in a 30 :_l C water bath for 30 min. Per cent transmittance through silica cuvettes was read at 490 nm with a Beckman DK-ZA spectro- photometer. A standard curve was made over the range of 0.0 to 0.2 mg carbo- hydrate using a standard solution of 0.1 mg galactose-mannose (1:1 w/w) per ml. Li id Total lipid was determined by modification of the procedure of Folch, Lees, and Sloane-Stanley (1957). One half to one gram samples were added with 100 m1 of chloroform-methanol (2:1 v/v) into 400 ml beakers and stirred for 30 min. One hundred milliliters of 2% (w/v) potassium chloride solution were added and stirring continued for 15 min. The mixtures were then centrifuged-International, model V, size 2, with rotor #226—-for 20 min at 1060 g. The lower lipid layers were removed with a syringe. Anhydrous sodium sulfate was added to remove any remaining residual moisture. The lipid extracts were filtered through Whatman No. 42 filter paper. TWenty-five or 50 ml aliquots were evaporated, first over steam and then in a 110 C hot air oven, in previously tared aluminum dishes. Control valves were determined for the solvent system. A§h_ Porcelain crucibles were heated in a "Hevi Duty" muffle furnace to 525 C, cooled, and weighed. One half to l g samples were placed in the tared crucibles and heated at 525 C until a.white ash formed (about 48 hr). The crucibles were cooled and weighed. Then the samples were re-ashed at 525 C (about 48 hr) and the crucibles again cooled and weighed. Moisture Total solids of fractions obtained during extractions were determined. Representative samples were weighed into tared aluminum dishes, which were dried to a constant weight (usually 4-24 hr) in a 75 C Blue Line air oven manufactured by Blue M Electric Co. 87 Moisture of the freeze-dried fractions was determined by a dif— ferent method. Two tenths to 0.4 g samples were weighed into small tared aluminum dishes. They were dried to constant weight over barium oxide in a desiccator under vacuum (one week). Total solids of the vole diets were determined by drying the samples to constant weight in a 110 C air oven (4-5 days). In all three cases the per cent loss in weight was reported as moisture. Saponin Analyses of the vale diets for saponin.was a modification of the method of Jones (1969). A two per cent human red blood cell solution was obtained from the local Red Cross. Two milliliters were added to a series of test tubes. About 100 mg of sample were added to eaCh tube. The tubes were shaken and allowed to stand. One hundred per cent hemolysis (indicated by a bright red color) within 30 min implied a very high level of saponin; and after 30 min a high level of saponin. If after 2 hr no hemolysis had occurred, the red blood cells settled to the bottom, indicating low level of saponin. Intermediate levels were qualitatively identified by degree of hemolysis as indicated by color; i.e., cherry red equaled medium high levels and pink medium low levels. The saponin analyses were performed by Miss Vicki Marcarian of the Department of Crop and Soil Sciences. Physical Methods High Voltage Peper Electrophoresis To evaluate the peptide content of the preparations, samples were Spotted in 50 ul concentrations (five 10 pl applications) on Whatman 3M paper. The paper was wetted in buffer of pyridine-acetic aciddwater 88 (1:10:189; v/v), pH 3.5. It was positioned in a laboratory made electro- phoretic apparatus of Plexiglass over a horizontal supporting bar with the ends in the electrode (Pt) compartments (which were filled with 500 ml of buffer). The buffer and hanging paper were overlaid with Varsol, which served as a cooling substance. The apparatus was connected to a Savant Instruments, Inc., high voltage source. The determination was made at 2000 V for 1 hr giving a current density of ca 5-6 ma/cm of paper width. The paper was then removed and dried in an air oven at 90 C. Next, it was sprayed with ninhydrin reagent (0.35 g ninhydrin, 14 ml collidine, 135 m1 acetic acid, 350 ml ethanol). Color was developed by replacing the paper in the air oven for 15 min. The pep- tides were identified by the purple color produced by their reaction with ninhydrin. Biological Methods Protein Efficiency_Ratio Composition of diets. The dry ingredients were weighed in the following proportions: protein source to give 7% protein vitamin diet fortification mixture 2% salt mix W 3% alpha-cell, hydrolyzed 20Z carbohydrate to bring total to 100Z The composition of the vitamin and mineral supplements is given in Tables 11 and 12, respectively. The carbohydrate mixture was made of 2 parts corn starch, 1 part sucrose, and 1 part dextrin. Vitamin free casein was used as the protein source for the control diet. The original design was to make sufficient diet for an intake of 6 g/vole/day for 6 days per replication, for six replications. However, Table 11. Composition of vitamin diet fortification mixture Vitamin* Amount (8) (mg) Vitamin A concentrate (200 units/g) Vitamin D concentrate (400 units/g) Alpha-tooopherol Ascorbic acid Inositol Choline chloride Menadione (K) p-aminobenzoic acid Niacin Riboflavin Pyridoxine hydrochloride Thiamine Calcium pantothenate U1 b wl-‘l-‘H-l-‘UINUIUIUIU'IOk OOOOMOSOOOONU‘I \l o Biotin 20.0 Folic acid 90.0 Vitamin B12 1.3 * Triturated in dextrose Table 12. Composition of salt mixture W Mineral salt Percentage Calcium carbonate 21.000 Capper sulfate ('5H20) 0.039 Ferric phosphate 1.470 Manganese sulfate (anhydrous) 0.020 Magnesium sulfate (anhydrous) 9.000 Potassium aluminum sulfate 0.009 Potassium chloride 12.000 Potassium dihydrogen phosphate 31.000 Potassium iodide 0.005 Sodium chloride 10.500 Sodium floride 0.057 Tricalcium phosphate 14.900 90 this was not possible due to inadequate amounts of certain protein frac- tions and actual formulations are given in Table 13. Table 13. Formulations of control and experimental diets for weanling meadow voles Ingredient Diet Casein 1P3* 183 2P3 283 3P3 383 Protein source 17.5 22.6 23.8 27.5 85.0 23.6 65.6 Vitamin mix 5.0 5.0 2.0 5.0 5.0 4.5 3.5 Salt mix 7.5 7.5 3.0 7.5 7.5 6.5 5.5 Alpha-cell 50.0 50.0 21.0 50.0 50.0 43.0 36.0 Carbohydrate 170.0 165.0 54.0 160.0 102.5 138.5 69.5 Total grams 250.0 250.1 103.8 250 0 250.0 216 0 180.1 Protein, Z 7.00 7.00 7.02 7 01 7.00 7.00 6 99 * See methods section, p 79, for explanation of nomenclature The dry ingredients were mixed with small amounts of water in a Hobart mixer, model N-50, to give a stiff dough. The dough was rolled out to a 1/4 in thickness and cut into rectangular 2 in by 6 in wafers which fit into the vale feeder (designed by Shenk and Elliott, 1969). Each such wafer was adequate for one animal for six days (one replica— tion). The wafers were dried in an air cabinet dryer (100-110 F) for 24 hours. They were then allowed to equilibrate for 24 hr to the relative humidity of the animal room. The diets were then kept frozen until 8-12 hr before a feeding trial was begun. At this time the wafers were allowed to thaw at room temperature. One gram of corn oil was spread over and absorbed by each wafer during this time. (Preferably the corn oil should have been added at a level of 2% in the initial stage of mixing the diets.) 91 Test animals. Entire litters of 6-7 weanling meadow voles (Miaratus pennsylvaniaus) were used as the test animal in this study. The vole colony is maintained by Dr. F. C. Elliott of the Department of Crap and Soil Sciences at Michigan State University in a room with a fairly con- stant temperature of 50 F and 24 hr light. Shortly prior to weaning and afterwards, but prior to feeding the experimental diets, the voles were fed an intermediate diet of the following composition: Milk (dried) 18% Carbohydrate 35% Alpha-cell 18% Butter or corn oil 5% Minerals 3% Vitamins 2% Honey 19% Experimental procedure. The voles were put on the experimental diets two to four days after weaning (15-17 days of age). Whenever possible, only those voles which had attained a weight of 11 g were used. The voles were randomly assigned to individual disposable plastic labora- tory cages with a crushed corn cob bed and some cotton. Each animal in a litter was fed a different experimental diet to reduce variance due to litter. The first two replicates (litters) included diets based on the acid precipitates of all three cuttings, the acid supernatants of the last two cuttings, and casein. The last five replicates included the above plus the diet based on the acid supernatant of the first cut- ting. The experimental diets and water were fed ad Zibitum for six days, with the exception of the seventh litter which was given the diets for only five days. Weight gain and feed intake were measured. (With the last replicate the voles were given the experimental diets for 24 hr before the initial vole and feeder weights were made. Total solids were determined on a sample of each diet so that intake on a dry basis could be calculated. Protein efficiency ratios were calculated as the weight 92 gain of the vale divided by his intake of protein (on a dry basis). Post-experimental Aut0psies The voles from the last two replications were sacrificed for autopsy. The first time the voles were examined for gross effects on the organs—~size, color, general appearance. The second time histo- pathological examinations of the intestinal tract, liver, kidney, and brain.were performed by the Veterinary Diagnostic Laboratories, Depart— ment of Pathology, Midhigan State University. Organoleptic Evaluation Cookie Formulation Twenty sugar cookies of each of the following formulations were prepared (see Table 14). The experimental fractions (previously stored one month) were substituted for flour in the control recipe. Table 14. Formulations of the control and experimental cookies for organoleptic evaluation Ingredient* Protein supplement Control Acid precipitate Acid supernatant Oleo 56.7 (1/4c) Sugar 99.1 (l/2c) Egg 27.4 (1/2) Sour cream 63.9 (l/4c) Baking soda 1.3 (l/4t) Baking powder 1.2 (l/4t) Salt 2.0 (l/4t) Nutmeg 0.33(l/8t) Vanilla 0.88(l/4t) Flour 148.0 (1 c) 142.15 133.4 118.8 125.5 56.2 Protein fraction --- 5.85 14.6 29.2 22.5 91.8 Alfalfa protein, Z 0.0 1.0 2.5 5.0 1.0 2.5 Alfalfa fraction, % 0.0 1.5 3.7 7.3 5.6 23.0 Flour substituted, Z 0.0 4.0 9.9 19.7 15.2 62.0 * amounts given in g 93 The cookie dough was prepared as follows: First the oleo and sugar were creamed. The egg was added and the mixture beaten until fluffy. Separately, the protein fraction, flour, baking powder, salt and nutmeg were mixed together. The baking soda was added to the sour cream. Alternately the dry and liquid ingredients were added to the creamed mixture. Lastly, the vanilla was added. The dough was refrig- erated overnight. Then it was rolled to 1/4 in thickness and 2-1/2 in diameter cookies cut and lightly sprinkled with sugar. The cookies were baked at 375 F for 12-1/2 min. Taste Panel Untrained panelists were asked to independently rate color, flavor, texture, and general acceptability of the cookies. A closed hedonic scale was used. It was assigned score points from 1-9 with 1 represent— ing dislike extremely and 9 like extremely. Figure 12 is a copy of the judging form given each panelist. Each cookie type was assigned a two— digit random number. For the first panel each of 20 panelists was given a plate with a control cookie and one each of the cookies sup- plemented with the three levels of acid precipitate. After rating these, each was given a second plate with a control cookie and one each of the cookies supplemented with two levels of the acid supernatant. A repeat panel of the acid precipitate supplemented cookies was con- ducted three days later with a second group of 20 panelists. Statistical Analysis of Data Mean values were calculated for each variable evaluated. Only results from the acid precipitate supplemented cookies and color values of the acid supernatant supplemented cookies were analyzed by analysis of variance (Amerine, Pangborn, and Roessler, 1965). Data showing 94 Name: Date: Plate No: INSTRUCTIONS: Evaluate color, flavor, texture, and general accepta— bility of each sample according to the appropriate hedonic rating scale. Record results in the table below. Color, flavor, texture and general acceptability Additional comments about any or all samples 9 like extremely 8 like very much 7 like moderately 6 like slightly 5 neither like nor dislike 4 dislike slightly 3 dislike moderately 2 dislike very much 1 dislike extremely RESULTS: color flavor texture general acceptability Sample score score score score Figure 12. Judging form for the organoleptic evaluation. 95 significant sample differences were further subjected to Duncan's multiple range test (LeClerg, 1957). RESULTS AND DISCUSSION Extraction Method The extraction procedure was intentionally kept relatively simple so that it might be extended to a large scale commercial operation. The alfalfa was Chopped to facilitate weighing and to create less of a strain on the blender. Also, water was added to accommodate the design of the blender. Singh (1967) noted that pre-chopping and extra.water added during extraction improved the efficiency and increased yields. Equipment specifically designed for leaf pulping (such as that described by Davys and Pirie, 1960) does not require the addition of water. Buf- fers were not used as they would be inconvenient for a commercial opera- tion. The blending time was kept short, for if extended to even 1.5 or 2 min the pulp becomes too fine and the fiber passes through the press slits into the juice. Slow warming of the dark green juice (about 5 C/3 min) caused coagulation and precipitation of the chloroplasts and thus removed most of the green color. If the juice was heated rapidly in a steameinjected hot water bath this effect did not occur. This initial removal of the chlorOphyll differentiates this procedure from the classical ones of Morrison and Pirie (1961) and Chayen at al. (1961). The juice was heated to temperatures below 50 C; for above 60 C there would be almost complete precipitation of the protein, resulting in a small yield of acid 96 97 precipitate at the final step. Continuous centrifugal filtration through Whatman No 2 filter paper was attempted but proved unsatisfactory for all centrifugation steps. Fine particles persistently passed through the filter paper and the larger particles rapidly plugged the pores. Supernatant l was clear and golden in appearance. The pH was adjusted to 8.5 (from about 5.7), because this has been claimed to increase yields (cf. Morrison and Pirie, 1961; Pirie, l966e). The last traces of green color were concentrated in the reSulting precipitate. Precipitation by acid was used initially because Mbrrison and Pirie (1961) recommended acid as the most convenient method for labora- tory-scale processing. In Modification A (p. 68) both heat- and acid- induced precipitation were used. Characteristics of the fractions produced are given in Table 15. Table 15. Characteristics of fractions prepared by Modification A Description Percentage Crude of Nitrogen of protein fractions (%)* original N (%)** Precipitate 3A (acid) 15.0 12.1 93.8 Supernatant 3A 4.4 18.8 27.2 Precipitate 3B (heat) 8.0 0.5 49.9 Supernatant 3B 4. 3 18 . 3 27 . 1 Precipitate 3A' (heat) 13.0 9.1 81.1 Supernatant 3A' 4.7 19.2 29.1 Precipitate 3B' (acid) 5.0 0.1 31.1 Supernatant 33' 1.1 19.9 25.9 *on a dry basis **per cent N x 6.25 98 Precipitate 3A, obtained by acid precipitation, contained 15% nitrogen. This constituted 12% of the original nitrogen. Further heat- induced precipitation resulted in another fraction containing 8% nitrogen-—only 0.5% of the original nitrogen. When the procedure of heat—induced precipitation was employed as a first step, a fraction con- taining 13% nitrogen was obtained which accounted for only 9% of the original nitrogen. The addition of acid to the clarified supernatant produced only a small amount of precipitate (5% nitrogen), accounting for but 0.1% of the original nitrogen. Most of the protein thus appeared to be labile to both heat and acid. This was expected as acid and/or heat have been used by previous investigators to induce precipitation of the protein. However, the acid method gave a more complete precipi- tation and a fraction of higher protein content. This is in agreement with the observation of Pirie (1952) that acid-induced precipitation resulted in a higher yield and purer protein isolate, and of Singh (1960) that more nitrogen was precipitated by TCA than by heat; but in con- trast to the latter's more recent reports (Singh, 1967, 1969). Morrison and Pirie (1961) reported that acid-induced precipitation denatures the protein less than heat-induced precipitation. Therefore, the procedure of acid-induced precipitation was adopted. Characteristics of Fractions Solubility. Unfortunately, the acid precipitate will not resolubil— ize upon neutralization in mildly alkaline solutions. This insolubility has been characteristic of leaf protein preparations (Pirie, 1957a, 1961a; Morrison and Pirie, 1961) and may be due to complex formation with tannins. According to Jennings et a1. (1968), Van Buren and Robinson (1969), and Loomis and Battaile (1969) tannins and other aromatic 99 compounds interact with proteins to form both soluble and insoluble complexes by reactions often involving polymerization of the phenols and covalent bond linkages to the protein. The nitrogen-containing components of the uncondensed supernatant fraction were not precipitated by TCA which, classically, would classify this fraction as non-protein nitrogen-—amino acids, small peptides or other low molecular weight nitrogen containing compounds. Yet, there was no loss of nitrogen during dialysis. A small portion was accounted for by glucosamines and/or galactosamines (from the amino acid chromato- grams), which possibly occur as moities of glyco-peptides or -proteins. The remainder was hypothesized to be peptides of molecular weight greater than 4000-6000, glycoproteins which are soluble in TCA or large NPN compounds (e.g., alkaloids). Concentration of the supernatant fraction resulted in the formation of a heavy flocculant which was readily dispersed by shaking, but which quickly settled. This precipitate did not form in the uncondensed super- natant when held overnight prior to condensing. Apparently, the phenomenon was a ramification of the solubility characteristics of proteinaceous components. Color and flavor. The acid precipitate was a cream colored and relatively bland (mild grassy odor and flavor) product. The color of the acid supernatant deepened as a result of concentration and appeared light brown in the 1y0philized form. It had a pronounced grassy odor and a strong bitter taste. Color photographs (taken against a white background) of the fractions obtained from the three consecutive cuttings of Field A are shown in Figure 13 (see p. 79 for nomenclature). (The blue-green overtone is artifact.) Color seems to be related to the 100 Figure 13. Acid precipitates and acid supernatants ob- tained by the standard extraction procedure from three cuttings. Figure 14. Acid precipitates obtained by Pirie extraction. lOl initial condition of the plant. The first cutting plants were green. The second and third cutting plants were yellow-brown; which produced a supernatant fraction that was a medium brown and a precipitate that appeared greyish. The color change was noticeable through the entire fractionation procedure. The juice from the latter two cuttings was yellowish-green instead of the deep green of the first cutting and all supernatants were brownishdyellow. Field B remained green through the entire growing season and the colors of the fractions obtained from all cuttings of this field were similar to those obtained from the first cutting of Field A. Solvent extraction (see Modification H, p. 74) of the freeze-dried acid precipitate and of the acid supernatant resulted in no observable color loss in either fraction, althoughthe discarded solvent from the acid precipitate had a slight yellowish color. Extraction with iso— prOpanol slightly reduced the "hay-like" flavor of the acid precipitate. However, there was no noticeable reduction in flavor of the supernatant, i.e., it retained its characteristically bitter flavor. Therefore, it was concluded that extraction of the protein isolates with isopropanol or acetone-hexane would not decrease the color or flavor defects suffi- ciently to warrant their use. The classical LPC was obtained by direct precipitation of the pro- tein from the juice (see Pirie extraction, p. 78), yielding a green powder which has received objections because of its color and grassy flavor (cf. Kamalamathan at aZ., 1969; Buchanan, 1969a). Four extrac- tions of this fraction with isoprOpanol resulted in a lighter colored precipitate, whereas an ethanol-hexane extraction did not remove the color as efficiently or completely. The precipitate, both before and after solvent extraction with iSOpropanol, is shown in Figure 14. 102 Compare these to 1P3 of Figure 13. Solvent extraction may prove to be an expensive process (cf. Pirie, l969d) and therefore, simple low-speed centrifugation was emphasized as a means for removing the colorébearing components. An attempt to remove the color by passing the supernatant over an ion exchange bed (see Modification A, p. 68) was unsuccessful because most of the protein remained on the column. This reduced the nitrogen contents of the supernatants 90% and the total solids contents 85% (see Table 16). Table 16. Characteristics of supernatant fractions before and after passage over an ion exchange bed Description Percentage Crude of Total solids Nitrogen of protein fractions (Z) (Z) (g) original N (%)** Supernatant 3B 1.40 4.3 0.39 18.3 27.1 Filtrate 0.23 1.7 0.03 1.4 10 8 Supernatant 33' 1.20 4.7 0.43 20.0 25.9 Filtrate' 0.20 2.7 0.05 2.1 16.9 *on a dry basis **per cent N x 6.25 Proteolytic Activity in the Juice Changes in NPN with time were measured both in the heated and unheated juice to determine if significant proteolytic activity occurred in the juice during extraction (see Modifications A and B, p. 68 and 69). Advantages of use of the NPN increment over other methods as an index of proteolysis have been discussed by Singh (1962). The results of these studies are presented in Table 17. 103 Table 17. Changes of alfalfa juice NPN from 0-24 hr after extraction Time N_1:N_* Modification A (Z) Modification B (hr) unheated heated unheated heated o 2.23 (4.65)** 2.73 (4.15) 1.16 (4.07) 1.46 (3.77) 0.5 1.21 (4.02) 1.51 (3.72) l 2.24 (4.64) 2.83 (4.05) 1.11 (4.12) 1.46 (3.77) 2 1.16 (4.07) 1.56 (3.67) 3 2.32 (4.56) 2.87 (4.01) 1.26 (3.97) 1.61 (3.62) 4 1.36 (3.87) 1.56 (3.67) 6 1.46 (3.77) 1.71 (3.52) 8 2.53 (4.35) 2.74 (4.14) 1.56 (3.67) 1.65 (3.58) 24 1.71 (3.52) 1.76 (3.47) nitrogen (Z) of the juice 6.88 5.23 * on a dry basis ** protein nitrogen given in parentheses In both assays the original NPN of the unheated juice was about 80% of that for the heated juice. Although the data indicated a small degree of initial proteolysis for the heated sample prepared by Modifi- cation A, there was no significant net increase in NPN over the 8 hr observation period. However, the unheated juice showed an increase of 4.4% in NPN (or decrease in protein nitrogen), relative to the total nitrogen content of the juice, after 8 hr. After about 2 hr an increase in the NPN content of the heated juice prepared by Modification B is observable. After 24 hr, an unrealistic holding time, the NPN content had increased only 5.7%. Unheated juice did not show an increase in NPN until after 3 hr. Thereafter, proteolysis increased at a faster rate than in the heated juice (see Figures 15 and 16). 104 IO '- ESP 6+- Modification B °/. 4__ . 2*- Modification A . 0 ' x L l 1 1 l l 1 A /\ | O 2 4 6 8 24 TIME, hr Figure 15. Increase in NPN as a percentage of the total nitrogen content of unheated alfalfa juice. °/. 4L X Modification B Modification A O O “9 .br— 0) CD If J) TIME, hr Figure 16. Increase in NPN as a percentage of the total nitrogen content of heated alfalfa juice. 105 Nazir and Shah (1966), who also investigated autolysis of unheated leaf extracts at room temperature, reported a loss in protein nitrogen of 3.5% in 2 hr and 26.5% after 24 hr. These values are over twice those obtained in this study. When working with large quantities of juice (about 12 l), the maximum time the unheated juice was held was 1 hr and, according to the results of the present study, there would be no detectable proteolysis. This agrees with the conclusion of Nazir and Shah (1966) that the juice can be kept up to 2 hr without signifi- cant protein destruction. Similarly, the heated juice was not held more than 1.5 hr. A second approach used to evaluate proteolytic activity was incu- bation with raw skim milk followed by high-voltage paper-electrophoresis (HVPE) (see Modification 1, p. 74). None of the tubes (milk plus juice) incubated at room temperature showed an increase in viscosity or clotting of the milk over the 6.5 hr time period. When incubated at 37 C, the milk sample containing 0.5 ml juice showed no coagulation, but the tubes with 1.0 and 2.0 ml of juice showed coagulation after 5 hr and 2.5 hr, respectively. This coagulation could be due to the proteolytic activity of the juice or to the complexing of the casein due to the presence of Ca+2 in the juice. In these two samples, no bands were observed by HVPE which were not present in the milk or juice controls (see Figure 17). Therefore, proteolysis of raw skim milk by fresh alfalfa juice could not be demonstrated by the appearance of peptides. Since extractions were conducted at room temperature, it was significant that no coagulation occurred in these experiments. Since the temperature was increased but once during the extraction (to 48 C), it was concluded that no significant proteolysis of leaf proteins occurred during the extraction procedure. C: 1' I 'th' C3 \ \ \ “I C: Alfalfa Milk + juice incubated at 37 C Milk whey (+ 2 ml juice) (+ 1 m1 juice) whey Figure 17. Schematic representation of high-voltage electrophoreto- gram showing peptide bands observed. Enzyme Extractions Cellulase. An attempt was made to increase the yields and purity of the final fractions by the use of carbohydrate hydrolyzing enzyme preparations. In the first experiment (Modification C, p. 71), Cellzyme was added during the blending step with the expectation that an increase in nitrogen content in supernatant 1 and a decrease in the nitrogen content of precipitate 1 would result. The results of this experiment are given in Table 18. During the eight hours of incubation there was little change in the total nitrogen of supernatant l extracted from the enzyme-treated leaf but there was an increase of about 9% in NPN. Although there was an increase of almost 30% in NPN of the control, its final protein nitrogen content was greater than that of the supernatant of the 107 Table 18. Nitrogen contents of fractions obtained after blending with Cellzyme * Fraction Time Nitrogen Non—protein Protein** (hr) (Z) nitrogen nitrogen (7.) (Z) Alfalfa 5.11 -—— ___ Juice control 0 6.01 ——— ___ Cellzyme 0 5.13 --— -__ Supernatant 1 control 0 4.47 2.03 2.44 Cellzyme 0 —-- 2.67 ___ Cellzyme 4 4.51 2.46 2.05 Cellzyme 6 4.43 2.65 1.78 Cellzyme 8 4.50 2.90 1.60 control 8 4.96 2.62 2.33 Precipitate 1 control 0 8.83 ——- __- Cellzyme 0 8 62 --— -_- Cellzyme 4 8.00 --— ___ Cellzyme 6 8.16 ——- __- Cellzyme 8 7.18 -—- ___ control 8 8.16 --- ___ * on a dry basis ** N Z - ZNPN enzyme-treated leaf. The compositional trends in the precipitates did not correspond to those observed in the supernatants. There appeared to be no increase in the protein nitrogen content of the supernatant of tne enzyme-treated preparation. Holding the pulp in a large scale Operation would not.be practical because (1) loss in protein due to adsorption and proteolytic activity is two to three times as high at this stage as in the juice (Davys and Pirie, 1969) and (2) eventually one machine will be designed to handle both pulping and pressing in one operation. For this reason, and since 108 the blending takes less than a minute, in the following four experiments with cellulase, amylase, and pectinase (Modifications D, E, F, and G; p. 73 and 74) the juice was extracted before incubation with the enzyme. Precipitate 3 was the major protein fraction and therefore, an increase in the nitrogen content of this fraction would be desirable. Nitrogen contents of the fractions isolated from the cellulase extractions are given in Table 19. Preparations obtained from the Table 19. Nitrogen contents of fractions obtained after incubation of alfalfa juice with cellulase Description Nitrogen* of (Z) fractions ** modification D modification E Alfalfa 4.31 3.78 Juice 5.14 4.61 control cellulase control cellulase Supernatant 1 t = 0 3.67 3.53 3.52 3.43 t = 4 3.79 3.51 t = 6 -—- -—- t = 8 3.85 3.83 3.55 3.83 Precipitate 1 t = 0 7.19 7.26 7.11 7.03 t - 8 7.86 7.25 6.11 6.65 Supernatant 3 t = 0 2.35 2.37 2.32 2.29 t = 8 2.49 2.24 2.66 2.69 Precipitate 3 t = 0 12.85 12.19 11.40 11.56 t 8 4 11.50 10.89 t = 6 11.10 --- t = 8 11.14 13.10 12.54 . 11.86 * on a dry basis ** modification D: +0.465 mg cellulase/lOO m1 juice modification E: +1.00 mg cellulase/lOO ml juice 109 enzyme-treated leaf material showed an increase in the nitrogen content of supernatant l. The data for precipitate 3 and supernatant 3 were inconclusive. In both experiments, there was a net increase with time in the nitrogen content of precipitate 3 (11.6-11.9Z and 12.2-13.1Z) obtained from the juice incubated with enzyme. However, at the lower enzyme concentration there was an increase in nitrogen of only 0.25% over that in the control preparation at time zero. At the higher enzyme concentration, precipitate 3 was lower in nitrogen than that of the control preparation after 8 hr of incubation and only slightly greater than that of the control at time zero. Apparently, treatment of the leaf material with cellulase enzyme preparations does not pro- duce increased protein purity in the leaf isolate. Amylase. The nitrogen contents of the fractions obtained from the extraction with alpha—amylase are given in Table 20. Relative to a normal extraction there was not a significant change in the nitrogen content of supernatant 1 from the enzyme-treated plant. Although there seemed to be a slight increase in the nitrogen content of precipitate 3 from the amylase-treated preparation after 8 hr of incubation (10.8- 11.7Z), the higher value was less than that of the control (12.1%). Pectinase. The results from the pectinase-treated leaf extraction are presented in Table 21. This enzyme-treated preparation showed a decrease in the nitrogen content of precipitate 3 from 11.6—10.0Z over the 8 hr of incubation. The control also showed a less pronounced decrease. Both preparations showed an increase in the nitrogen content of supernatant 3 during the 8 hr incubation period. Thus, there appeared to be some proteolytic activity in both preparations. 110 Table 20. Nitrogen contents of fractions obtained after incubation of alfalfa juice with alpha—amylase Description Nitrogen of (Z) fractions control o-amylase Alfalfa 2.74 Juice 4,04 Supernatant 1 t = 0 3.27 3.29 t = 4 3.23 t = 6 3.34 t = 8 3.38 3.31 Precipitate l t = 0 6.00 6.64 t = 8 6.24 6.77 Supernatant l t = 0 2.32 2.28 t = 8 2.60 2.64 Precipitate 1 t = 0 12.11 10.84 t = l. 11.10 t = 6 9.64 t = 8 12.07 11.68 At the enzyme levels and conditions (room temperature and pH about 5.7) employed, there was no evidence that treatment of leaf materials with the various enzymes (cellulase, alpha-amylase, and pectinase) enhanced the purity of the protein recovered. However, because the amounts of protein recovered as well as its purity are of interest, more relevant data would have been realized if fractional yields had been recorded. 111 Table 21. Nitrogen contents of fractions obtained after incubation of alfalfa juice with pectinase * Description Nitrogen of (%) fractions control pectinase Alfalfa 3.60 Juice 4.28 Supernatant l t - 0 3.50 3.53 t = 4 __._ t = 6 3.45 t = 8 3.74 3.73 Precipitate 1 t = 0 6.97 6.84 t = 8 6.98 6.09 Supernatant 3 t = 0 2.53 2.61 t = 8 2.80 2.93 Precipitate 3 t = 0 11.74 11.64 t = 4 11.31 t = 6 10.89 t = 8 10.97 10.00 * on a dry basis Yields Greenhouse alfalfa. Yield studies, based on nitrogen (N) and total solids (TS) determinations (see p. 77) were made of three extractions of greenhouse alfalfa: January 6, 1970 (1); February 17, 1970 (II); and April 6, 1970 (III). The results of these studies are recorded in Tables 22, 23, and 24. The nitrogen content of the fresh alfalfa ranged from 2.8% (II) to 3.6Z (III) to 4.5% (I). This range is close to that reported for alfalfa by Guggolz, Herring, and Kohler (1967). The total solids content decreased with corresponding increases in nitrogen con- tent (i.e., 25.3% TS and 2.8% N to 16.3% TS and 4.5% N). This negative 112 Am unnumaquva unnumoummsm wwom mm mmammam ova nH\awmuoua hp Am oumufiawoauav mumuamaomua have no mmammam uo3 nH\ofimuoua hp 6 w m: 6 w mm.N "samew 2.0 x 2 as wanna. be m co «m o.H m.mm m.m~ H.NH om.o Ho.mH e.m N.H o.mH m.o m.ma m mumuwafiomum m.N N.n~ n.qm w.wH o¢.o mm.q m.ma m.m o.H ~.qo mmm m ucmumaumasm N.o «.ma q.~ m.a no.0 HH.m ¢.H m.o w.m o.a w.m~ N mumufiafiomum w.m o.mm «.mm «.mm Ho.o mm.m o.HN n.0H H.H ~.oo mom N acoumchmasm m.m m.Hm N.oq w.HN no.0 om.m m.HH o.m q.m H.n m.moa H mumufimfiumum H.q q.om m.om 0.0m 00.0 ~w.m m.m~ m.HH H.H «.05 «moa H unnumohmasm 0.x o.mNH msoflmau Hanan m.m o.m¢ 0.00H N.qm 0H.H mw.o m.mm m.oH q.H m.Hm HmHH mawsh 0.00H o.nmqa umumB + mwamwa< ooHH swam: «.ma H.mm III o.ooa qa.m mq.q o.ooa n.mq m.oa o.Hm~ mmammam va kxANv z manmuomuuxm z Hmswwfluo va xANv mH Hmafiwfiuo Amv ANV us Hmcfiwfiuo va maowuomum mo mo mo mo mo afimuoum.mosuo ammuamoumm ammucmoumm cmmouufiz ammuomoumm moHHom HmuOH ammuomoumm unwwmz coaumwuamaa aH nowuomuuxm mm ou omuuommuv monuma cowuumuuxm oumvcmum mnu mo < cowumoHMHooz :uHB mauamufiaoo Iooo omuosoaoa moose mama» ma omaHMuno mGOfiuamum mama mmammam mo moaumaumuomumna Hmoofiuwmoaaou .NN manmfi 113 Am unmumcummsmv unnumcuoasm oHom mm memem um: nH\aHmuoua mun Am mumuHaHomuav muMuHmHomua mHom mm NMHmMHm umB nH\aHououm mun» m: N 00. H mHmmn mun m so "onHw mN.o x 2 um um m.o w.oo w.mH o.m o.H Nn.m q.H q.OH m.NH o.o H.Hm m mumuHmHoaum o.nH o.NH o.wm w.mH w.N Nm.H c.0N N.N«H m.H m.no whom m ucmumauaasm m.H m.m N.N o.H N.o om.H w.H o.mH ¢.m H.H m.mmH N mumuHmHooum c.¢N H.0H N.mm m.mH m.m mm.N H.HN o.NmH o.H o.mo memo N uqmumoumasm o.mH q.om N.Nq «.mH H.m Nm.m 6.5 m.mm H.OH N.m 0.0mm H mumuHaHumum m.wN m.NH N.Ho N.NN m.q ow.N N.NN N.H©H o.H w.HN meOH H unnumcumasm H.¢H o.HmON msonou umnHm N.oq N.NN 0.00H m.om «.5 mm.m N.wN o.wON m.H m.mm mQMHH moHsh 0.00H m.quqH umumB + mmHme< ¢.mquH umumz m.NNH o.NH III o.OOH c.0N Hw.N o.OOH w.an m.mN H.womN mmHme< va «RANV z oHomuomuuxm z HmonHuo va «ANV we HmanHuo Amy ANV us HmcHwHuo va mGOHuumum mo NO NO mo mo chuo m mosuo ammuomoumm ammucmoumm omwouqu ammucmaumm moHHom HmuoH ammusaaumm uanms :OHunHuommo AHH aOHuamuuxm mm on omuummauv panama GOHuomuuxm pummcmum was no mwsum onHm an omchuno mGOHuomnm wmoH mMHmMHm mo moHumHuouomumno HmQOHuHmomBoo .mN mHomH 114 A0 unaumcummsmv ucmumouoaom 0Hom mm mMHmMHm umB 0H\sHmuou0 0H0 A0 mumuHaHomuav mumanHomum 0Hom mm memmHm ums 0H\onuoum how 0 w m HN.0 HN.0 "vaHw 0N.0 x z «« Hmmo 0H0 m so . « H.5 0.50 5.0N 0.0H H.H 05.0H 0.0 0.0H N.0H H.H 0.00 0 muMuHaHomum H.5 H.0H 5.0N 0.0H H.H N0.N 0.HN 0.00 0.H N.N5 0000 0 uamumsumasm ~.H m.n~ m.e 0.N ~.o ,mo.0 m.~ H.m m.m m.H N.Nm N eemuseeumee 0.NH 0.0N 0.00 H.0N 0.N 00.0 0.0N N.N0 0.H 5.05 0500 N uamumoumasm 0.NH 0.N0 0.00 0.0N 0.N 00.0 0.0H 0.0N 0.0 0.0 0.000 H mumuHmHumum 0.0H 0.0N H.N0 0.00 0.N 00.0 0.0N 0.00 0.H 5.55 0500 H uomumoumasm 0.HH H.550 mstmmu umnHm 0.5N 5.N0 0.00H 0.00 0.0 0N.0 0.00 0.00 0.N 0.00 0HNO 00H30 0.00H 0.0000 umum3 + mMHm0H< . N000 swam: 5.N0 0.NN III 0.00H 0.0 H0.0 0.00H 0.5HN 0.HN 0.500 MMHm0H< Amy «xANV z mHnmuomuuxa z HmcHano va «AN0 05 HmanHuo A00 5N0 us HmonHuo H00 mGOHuomum mo 00 00 mo 00 chuoumlmvsuu ammuomoumm ammuomoumm camouqu ammuomaumm mvHHom HmuOH mwmusmoumm uLMHms GOHumeammo AHHH aOHuomuuxm mm Ou wouuammuv wosume :OHuomuuxm wumvomum ecu mo 0 SOHumoHMHuoz :uHB zHuGMuHEoo Iaoo wmuusvaoo zpsum UHMHm 00 vaaHmuno mGOHuamuw mmmH mMHmMHm mo maHumHumuomumso HMGOHuHmoaaou .0N mHQmH 115 correlation has been reported between protein and fiber contents at all stages of growth (Van Riper and Smith, 1959; Ogden and Kehr, 1968). The low nitrogen specimen represented plants which had reached the seeding stage. This is in agreement with reports that the protein content of alfalfa decreases with maturity (Van Riper and Smith, 1959; Ogden and Kehr, 1968). Only 36% of the leaf nitrogen was extracted from the more mature specimen. In contrast juice from the young plants contained more than 50% of the original leaf protein. That more protein can be extracted from young than from mature plants has been well documented (cf. Pirie, 1962; Singh, 1964; Nazir and Shah, 1966). However, specimen 111, containing 3.6% nitrogen showed greater extractability than specimen 1 which had a higher nitrogen content (4.5%). This is in agreement with recent work indicating that high extractability and high nitrogen content in the leaf are not necessarily positively correlated (Byers et aZ., 1956b; Byers and Sturrock, 1965; Nazir and Shah, 1966). The nitrogen content of the juice varied from 3.6-6.9% reflecting the nitrogen con- tent of the leaf. Precipitate 1 (chloroplastic) contained from 5.8-8.3% nitrogen, which also increased with the ritrogen content of the leaf. This frac- tion represented about 15-30% of the original leaf nitrogen or slightly less than half of the extracted nitrogen. The remainder of the extracted nitrogen remained as soluble nitrogen in supernatant 1. Here again, the supernatant nitrogen content varied, increasing with increasing nitrogen content of the leaf. Only small amounts (1.0—1.3%) of the leaf nitrogen appeared in precipitate 2 (alkali-induced), containing from 1.6-4.1% nitrogen. Thus, supernatant 2, with nitrogen contents of 2.6-5.9%, retained from 116 19-28% of the original nitrogen in a soluble form. This soluble nitrogen fraction was split between the two final fractions, i.e., acid precipitate and acid supernatant. About half, representing from 14—19% of the leaf protein, remained in the acid soluble supernatant (supernatant 3) which contained from 1.9-4.4% nitrogen. The precipitated fraction (precipitate 3) accounted for 5.0-16.5% of the leaf nitrogen. The low yield (5.0%) was obtained from the older plant extract. The nitrogen content of this fraction, representing the major protein fraction, varied from 9.7-15.0%. Thus at the conclusion of the procedure, about 25% of the nitrogen extracted from young plants appeared in the major protein fraction (acid precipitate), whereas about 30% remained in the acid supernatant. The higher the nitrogen content of the leaf the greater was the nitrogen content of all resulting fractions (except for precipitate 2). However, the yield of the original leaf nitrogen in each fraction (except for supernatant 3) increased with greater extractability of the leaf-—extractability, as noted previously, did not always correlate with leaf nitrogen if other conditions, such as age, were similar. Consequently fractions obtained from extraction 11 (mature plants) always showed the lowest nitrogen contents and lowest yields of original nitrogen. Nitrogen contents of fractions from these plants were only about 45-75% as great as, and yields of leaf nitrogen only 35-80% that of, those from young plants. With one exception, fractions from extrac- tion I showed the highest nitrogen contents, whereas fractions from extraction III showed the highest yields of original nitrogen. All three supernatants from older plants contained a higher yield of extracted nitrogen (110-125%) than the same fractions from younger 117 plants. However, precipitate 3 (comparable to the various cyt0plasmic fractions reported in the literature) from the older plants contained only half as much extracted nitrogen as that from the young plants. This is in contrast to Pirie (1963, 1964b) who reported that the chloro- plastic to cytoplasmic protein ratio does not vary with maturity, and to Henry and Ford (1965) who claimed that mature leaves contain a higher proportion of cyt0p1asmic proteins because the reduction in extractable protein with maturity affects the chloroplastic more than the cytoplasmic fraction. Summarizing, the young plants yielded 2.5-3.2 g dry protein as acid precipitate and 3.2-3.85 g dry protein as acid supernatant per lb of wet raw material. Plants in the seeding stage yielded only 1.0 g and 2.8 g dry protein as acid precipitate and supernatant, respectively, per 1b wet alfalfa. Field alfalfa. The results obtained from yield studies made during the summer field harvests are given in the following tables: Table 25. First harvest of the first cutting (6/5/70) Table 26. Fifth harvest of the first cutting (6/11/70) Table 27. Second harvest of the second cutting (7/18/70) Table 28. Fifth harvest of the second cutting (7/22/70) Table 29. Third cutting (10/4/70) The first cutting showed the typical negative correlation between nitro- gen (5.2% and 3.0%) and total solids (15.6% and 24.3%). The crop condi- tion (maturity, greenness, leafiness) was not visibly distinguishable between the two harvests, yet there was a noticeable difference in leaf nitrogen. Both specimens showed extractabilities greater than 50%; the lower nitrogen leaf having the greater extractability (62%). As previously noted in the experiments with greenhouse alfalfa, for all fractions except precipitate 2, the higher the nitrogen content of the leaf the 118 A0 uamumoummsmv uomumoumasm 0Hum mm mMHmMHm ums 0H\nHouou0 mow 0 00.0 a0 oumuHmHaaumv oumuHaHomua 0H0m mm mMHmMHm nos 0H\onuoum 0H0 0 00.0 "0Hme 0N.0 x 2 xx mHmmn 0H0 m do an 5.0H 0.N5 0.0N 0.0H 5.N 0.HH 0.0 0.NN 5.0H N.H 0.50H 0 mumuHaHomum H.0N 0.0N 0.N0 0.0H N.0 00.0 0.0N 0.00 H.H 0.00 0000 0 acoumosmasm 0.H 5.0H 0.0 0.H 0.0 0H.0 0.N N.0 5.0 0.H H.NOH N mumuHaHomum H.00 H.H0 H.00 0.0N 0.0 50.0 0.00 0.00H N.H 5.H0 00H0 N unaumstmsm 0.0N 0.N0 0.00 0.NN N.0 05.0 N.5H 0.N0 0.0H H.0 0.000 H oumuHaHomum H.00 H.00 0.00 5.0N 0.0 H0.0 0.H0 N.0HH N.H H.00 0500 H uamumcumasm 0.5 0.000 msvamu umnHm 0.00 0.00 0.00H 0.H0 5.0 05.0 5.00 H.00H 5.H 0.00 0500 moH20 0.00H NONHH seams + mMHm0H< 0500 umumz 5.0HH 0.N0 III 0.00H 0.0H 0N.0 0.00H H.N00 0.0H NONN mmHmMH< A00 «eANV z mHomuomuuxm z HmcHwHuo A00 «ANV 0H HmcHwHuo 500 5N0 us HmonHuo A00 wGOHuomuw mo 00 mo 00 mo onuoum,mvsu0 ommuomoumm owmuomoumm cowouqu mwmucooumm mvHHom HmuOH mwmuamouom uanmB GOHuaHuomoQ mMHmmHm 0HmHm mo 0cHuu=o umuHm map 00 umo>umn umuHm mnu Scum wonuma GOHuomuuxm wumcomum ago 00 mwsum 0HmH0 00 wmchuoo chHuomum wmmH mMHmMHm mo moHumHumuamumno HmcoHuHmanou .0N mHan 119 A0 uamumcummsmv unnumohmmsm 0Hom mm NMHmMHm umB 0H\aHmuoum 0H0 A0 mumuHmHooumv oumuHaHuoua 0Hom mm mmHmMHm nos 0H\sHmuoum 0H0 m $.m m No.0 mHmmn 0H0 m do uvaHw 0N.0 x z 0* an 0.0H H.05 H.H0 0.0H N.0 0.NH 5.0 0.0N 5.0H N.H 0.00H 0 mumuHmHooum 0.5H H.5H H.0N 0.5H 0.N 05.N 0.0H H.00H N.H 0.05 0N00 0 uamumasmasm N.N 0.0N 0.0 H.N 0.0 00.0 5.H N.0 0.5 H.H H.HNH .~0.mumuHmHomum 0.00 0.0N 0.50 0.0N 0.0 00.0 0.NN 0.HNH 0.H 0.05 00H0 N uamumoummom 0.0N N.00 0.00 0.0N 0.0 50.0 0.0H 0.00 0.0H 0.0 0.0H0 H mumuHmHomum H.N0 0.0N 0.00 0.H0 H.0 N0.0 0.0N H.00H 5.H 0.00 N000 H uomumcumasm N.0H 0.NOHH msUHmmu HmnHm 5.00 0.H0 0.00H 0.N0 N.0H 00.0 0.50 0.00N 0.N 0.00 0000 ooH90 0.00H 000HH soumB + mMHmMH< 0000 woumz 0.NOH 0.0H III 0.00H 0.0H H0.0 0.00H H.000 0.0N 00NN m0H00H< A00 *«AN0 2 aHowuamuuxw z HmaHmHuo A00 *AN0 09 HmanHuo Amy ANV as HmchHuo A00 moOHuumum mo 00 mo mo mo chuo m mvsuo ammuomoumm mmmucmoumm ammouqu mwmucmoumm mvHHom HmuOH m0Mucmonom uszmz GOHuaHuomma mMHmem 0Hon 00 waHuuso umuHm mnu mo umm>um£ aumHm mnu Scum wonume GOHuomuuxm vumwomum mnu mo mvsum 0HmH0 00 vmonuno mGOHuomum mmoH mMHwMHm mo moHumHumuomumzo HmcoHuHmoQEOU .0N oHpmH 120 higher the nitrogen content of the resulting fractions. There was one noteworthy exception; precipitate 3 from the 3.0% nitrogen leaf con- tained 12.5% nitrogen, whereas the same fraction from the 5.3% nitrogen leaf contained only 11.6% nitrogen. In all cases, each fraction from the lower nitrogen leaf represented a greater yield of the original nitrogen due to the enhanced extractability of this specimen. This latter harvest also showed slightly higher percentages of extracted nitrogen in the precipitates and lower percentages in the supernatants than observed for the first harvest. The nitrogen contents of the various fractions were in the same range as those obtained from greenhouse plants at about the same stage of maturity (i.e., extractions I and III) although the average nitrogen contents of the "greenhouse" fractions were slightly greater than the "field" fractions. The "field" and "greenhouse" fractions also showed similar distributions of the extracted nitrogen. There was one dif- ference; namely, field alfalfa precipitate 3 contained a greater yield of the leaf nitrogen (or extracted nitrogen) than the same fraction obtained from the greenhouse alfalfa--17.0% (or 29%) as compared to 14.3% (or 24%), respectively. Consequently, field harvests gave higher yields of protein in the acid precipitates averaging 3.65 g dry protein per 1b wet alfalfa (greenhouse yields averaged 2.85 g per lb). There was little difference in the yield of acid supernatant protein between the two sets of plants. The second cutting was lower in nitrogen (2.5% and 2.2% for second and fifth harvests, respectively) and higher in total solids (32.4% and 32.3%) than the first cutting. This was expected from the condition of the plants. Although they were the same plants at the same stage of maturity, the regrowth was yellowish, dry, and stemmy, whereas the first 121 A0 unmumoummomv usmumoumasm 0Hom mm mMHmem ums 0H\chuoua 0H0 0 0 A0 mumuHmHomumv mumuHaHomum 0Hom mm memMHm umB 0H\chuou0 huv 0 N U $0 mHmmn 0M0 m so "vaHw 0N.0 x z «x K N.0 0.00 0.HN 0.0 0.H 50.0 N.N 0.0H 0.0H 0.H 0.50 0 oumuHmHomum 0.0H 0.0H 0.00 0.0H 0.N 00.N 0.0H 0.00H 0.H 0.00 N050 0 oomumcum050 H.N 0.NN 0.0 H.N 0.0 00.0 0.H N.0 0.5 0.H 5.HOH N mumuHaHooum 0.HN 0.0H 0.50 5.HN 0.0 00.N 0.0H 0.0HH 5.H N.05 0000 N ucmumcumasm H.0 5.0H 0.0N N.0 0.H 0H.N 0.0 0.00 0.00 5.H N.00H H mumuHoHomum 0.0N 0.0H 0.00 0.0N 0.0 0H.0 0.0H 0.0HH 5.H 0.N5 50N5 H uqmumouomsm 0.NH 0.00NH umon 0.50 0.0H 0.00H 0.50 0.0 0H.0 N.00 5.00H 0.N 0.05 5005 moHsh 0.00H 0.0000 poems + mmHmmH< 0005 noumz N.00 0.0H III 0.00H 0.0H N0.N 0.00H H.0N0 0.N0 0.000H mMHm0H< Amv AN0 2 MHnmoomuuxm z HmcHwHHo va ANV 0H HmonHuo A00 ANV us HmcHwHuo A00 mGOHuamum «« x 00 mo 00 wo wo onuoumlmvsuo mwmusmaumm mwmuomouom omwouqu ammucaaumm mvHHom Houoe mwmuamoumm usmHmB aOHuQHuomma mmHmMHm 0Hon mo moHuuso 0coomm msu mo umm>um£ 0coomm mnu Baum 000008 GOHuamuuxm wumwomum msu mo 005um 0HmH0 an noaHmuno mGOHuomum 0mmH mMHmMHm mo moHumHumuomumno HmaOHuHmomeoo .5N mHan 122 Am unaumoummsmv ucmumnummsm 0Hom mm mmHmMHm ums 0H\aHmuoum 0s A0 mumuHmHoouav aumuHmHamum UHUN mm MMHmMHm uaB 0H\chuoum 0p m e me e m Hm. mHmmn 0H0 m co m H "eeees 0N.0 x z «a an H.0 H.00 0.0N 0.0 0.H 0N.0H 0.H 0.NH 0.HH H.H N.00H 0 muMuHmHomum H.0H 0.NH 0.N0 0.5H 0.N H0.N 5.0H H.0NH 0.H 0.N5 00N5 0 uamumcumasm 0.N 0.NN 0.0 N.N 0.0 00.0 0.H 0.0 N.5 N.H 0.0NH N mumuHmHumum 0.NN 0.0H 0.05 0.0N 5.0 00.N 0.0N 0.00H 0.H 0.05 0N05 N usmumnum050 H.0H 0.0H 0.N0 0.5H 0.N 00.N 0.0H 0.00H 0.00 0.0 H.000 H mumuHmHomsm 0.0N N.5H N.05 0.0N 0.0 05.N 5.HN N.HOH 0.H 0.05 0005 H oomumcnmmsm H.0H 0.0NOH osvamu umnHm H.H0 0.5H 0.00H 0.00 0.0 05.N 0.5N 0.05H N.N 0.H0 H5H0 maH50 0.00H 0.5000H Rooms + mMHm0H0 0500 umumz N.00 0.0H III 0.00H 0.0H NN.N 0.00H 0.000 0.N0 0.5HON mMHm0H< Amv eeANv z maneueeeuxm z Heaewsuo Amv «ANV my Hmeewseo Amy ANV us Hmesmeeo Amv eeoeuueee mo 00 mo mo 00 onuon0,mwsu0 ammuomouom ammuomoumm cowouqu mwmuomonmm mvHHom HmuOH mwmucmouom uanmz GOHuaHuommo mmHmmHm 0HmHm mo 0cHuusu cacaom mnu mo umm>umn nuMHm mnu Baum wonume ooHuomuuxm vumwomum oeu mo zwsum 0HoH0 00 0maHmu00 mGOHuamum mmmH MMHmMHm mo moHumHumuomumso HMGOHuHmOQEOU .0N mHan 123 growth was green, leafy, and tender. Extractability of these regrowths was low: 34% and 38%. The second harvest had a slightly greater (0.3-0.4Z) nitrogen content in the leaf, juice, and three supernatants than the fifth harvest. However, the acid precipitate contained only 9.4Z nitrogen as compared to 10.3% nitrogen in this same fraction obtained from the fifth harvest. The nitrogen contents of all fractions extracted from the second cutting, except precipitate 2, were significantly less (generally about 40-80%) than those observed for the same fractions obtained during the first cutting. Data of Hartman at al. (1967) indi- cated that the second cutting alfalfa and the juice extracted from it had protein contents only 90% of those of the first cutting alfalfa and resulting juice. Earlier Tilley et al. (1954) had reported second cutting alfalfa to have a higher crude protein content than the first cutting; although, the crude protein content of the LPC from the second cutting was lower than that from the first. The lower percentages of leaf nitrogen in the second cutting frac- tions were expected because of the lower extractability encountered. Particularly striking was the observation that although supernatant 3 retained almost the normal amount of leaf nitrogen (16%), the acid pre- cipitate contained only 8.6%. A comparable yield from the first cutting was 17Z. Evidently, the reduction in leaf nitrogen in the plant con- sisted primarily of a reduction in the chloroplastic protein--as might be expected by the yellowish color of the plants--and in the amount of protein precipitated by acid. On the basis of the percentage of extracted nitrogen the percentages for the supernatant of the second cutting are greater by 25-50% than those found in the first cutting; although the percentages for precipitates l (chloroplastic) and 3 (acid) were 50% and 20% less, reapectively. This increase in the ratio of soluble to 124 precipitable protein was also noted in the older greenhouse alfalfa which exhibited a low nitrogen content and low extractability. Whereas the age difference of the greenhouse alfalfa caused the greatest reduc- tion to occur in the acid precipitable protein, differences in color and lushness of the field alfalfa resulted in the largest reduction in the chloroplastic protein. The plants of the third cutting were similar in appearance to those of the second cutting, i.e., they were brown and dry. Nitrogen (2.5%) and total solids (31.7%) contents of the leaf were not greatly different from the second cutting. Nitrogen contents of the juice and precipitate l were slightly greater than, supernatant 3 and precipitate 3 within the range of, and supernatants 1 and 2 slightly less than, those of the second cutting. The extractability--50%-—was much higher than the second cutting although slightly less than the first cutting. However, the percentages of leaf nitrogen in all extracted fractions were not necessarily greater than those of the second cutting. All three supernatants were within the same range as the second cutting; the increased extractability being exemplified by an increase in the recovery of leaf nitrogen in precipitates 1 and 3. A comparison revealed that percentages of extracted nitrogen in the third cutting supernatant fractions were significantly lower than those of the second cutting, whereas percentages for precipitates 3 were about equal and the percentage for precipitate 1 from the third cutting was about twice that of the second cutting. Thus, except for precipitate 3, the third cutting showed a distribution of extracted nitrogen similar to the first cutting, although the nitrogen contents of the fractions obtained were not much different than those for the second cutting. It should be remembered that all three cuttings were 125 A0 oomumouomsmv unaumnummsm 0Ham mm mMHmMHm ums 0H\:Hmuou0 0H0 0 00.0 A0 mumuHmHomuav mumanHomum 0Hom mm mmHmmHm nos 0H\:Hmuoum huv 0 H0.N "0Hme 0N.0 x z as mHmmn 0H0 m so . a 0.0 0.N0 N.NN H.HH 0.H 00.0 0.N 0.0H 0.NH 0.H 0.0NH 0 mumuHaHomum 0.0H 0.HH H.N0 H.0H N.N 00.H 0.HN 0.5HH 0.H 0.05 00N0 0 uomumoummsm 0.H 0.0H H.0 0.H N.0 H0.0 0.H 0.5 0.5 N.H 0.00 N mumuHmHomum 0.0N 0.0H 0.00 N.0N 0.0 00.N 0.0N 5.0NH 0.H 0.05 H000 N oomumsummsm 0.5H 0.0N 5.H0 0.0N 0.N 05.0 0.0H 0.05 0.NN 0.0 0.000 H oumuHmHumum 0.HN 0.0H 0.00 N.0N 0.0 00.N 0.0N 5.0NH 0.N 0.05 0000 H uomumsummsm 0.0H N.50HH msvaou umnHm 0.N0 0.HN 0.00H H.00 0.0 H0.0 0.00 0.HON 0.N H.N0 5005 moHsh 0.00H 0.NH00 umums + meme< 0000 umumz 5.00 5.0H In: 0.00H 5.0H H0.N 0.00H 5.000 5.H0 0.NN5H mem0H< va *4AN0 z maemuumuuxm z Hmesmseo va eawv my Haeemseo va ANV us Hmesmeeo Amv meoeuumee mo 00 00 mo mo onuoum mvsuo mwmuomouom ammunmoumm amwouqu ammucmoumm meHom HmuOH mmmucmouam usmHmB GOHuaHuome «HHmem 0HmHm mo 0sHuuso quau mau Baum wonume GOHuomuuxm wumwcmum mnu mo mwsum 0HmH0 00 woonuno mGOHuumum 0mmH mMHmMHm mo moHumHumuomumno Hm:0HuHm00800 .0N mHan 126 made at as nearly as possible the same stage of maturity (using the criteria of per cent bloom). Pirie (l966e) reported a study of regrowth patterns on the yield of extractable protein that gave conflicting results. The nitrogen contents of fractions obtained from the greenhouse plants gathered at the post-seed stage were within the range of those of fractions extracted from the last two cuttings. Exceptions to this trend were encountered in the juice and the chloroplastic precipitate which contained higher contents of nitrogen, reflecting the green color of the greenhouse plants. The percentages of leaf nitrogen in the fractions were also of the same magnitude as those of the second and third cuttings. As noted previously the two groups of plants showed a similar extracted nitrogen distribution with the exception of the acid and chlorOplastic precipitates. Total solids. The T8 content of the alfalfa varied inversely with the nitrogen content of the leaf, ranging from 15.6-32.4Z. Generally, the TS content of the isolated fractions were proportional to the T8 of the leaf. It was also noticed that (1) there was a gradual decrease of about 0.7% in the TS from the juice down to supernatant 3 and that, (2) of the precipitates, precipitate 2 was always lowest in T8 and either precipitate 3 or precipitate 1 would be highest. Ranges of TS content of the various fractions obtained from the several preparations are listed in Table 30. About 65% of the leaf solids was retained in the fiber residue. Another 17% was precipitated with the chloroplastic fractions, leaving about one-quarter of the leaf solids in supernatant l. The alkali-induced precipitate contained slightly less than 2% of the original solids. On the average, the final 127 Table 30. Total solids characteristics of fractions extracted from alfalfa Description of Total solids Percentage of fractions (Z) original TS Alfalfa 15.6 - 32.4 100.0 Juice 1.4 — .2.0 27.5 — 46.7 Supernatant 1 1.1 - 2.0 19.0 - 31.3 Precipitate 1 5.4 - 30.5 7.4 - 17.2 Supernatant 2 1.1 - 1.9 18.4 - 30.4 Precipitate 2 3.8 - 8.4 1.3 - 2.5 Supernatant 3 1.0 - 1.9 16.0 - 26.5 Precipitate 3 11.6 - 19.2 1.4 - 6.3 supernatant and precipitate accounted for 20.5% and 3.5%, respectively, of the leaf solids. This can be compared to 16.5% and 12%, respectively, of the leaf nitrogen. There were no particularly striking differences in the distribution of solids between the various preparations. Summary. The extraction data seemed to fall roughly into two groups: (1) from plants in "good" condition--young and green (extractions I and 111, first cutting), and (2) from plants in "poor" condition--mature or brown (extraction 11, second and third cuttings). Some of the pertinent nitrogen data are recorded in Table 31. The previously reported crude protein contents of fresh alfalfa from which LPC was prepared have ranged from 19.5—33.1z* (Valli Devi et aZ., 1965; Smith, 1966; Hartman at aZ., 1967; Subba Rau et aZ., 1969; Spencer et aZ., 1969; Knuckles at aZ., 1970), which is comparable to our "good" alfalfa. The average protein content of juice extracted from "good" alfalfa is within the range of 28.0-36.3% protein reported by Hartman et al. (1967), Subba Rau et al. (1969), * Throughout the thesis compositional data (both from the literature and this study) will be given on a dry basis. 128 00.0 50.0 0 acmumcummsm Hm.H m~.m m mumeeeeemnm Moon 0000 0H\0 "m0Hme 0N.0 x z «« mHmmn 0M0 m :0 « 0.H0 H.05 0.HN 0.0N 0.0 0.0H 00.0 0.NH 0 mumuHmHomum 0.NH H.0N H.00 0.00 0.0H 0.5H 00.N HN.0 0 00000090000 0.0H H.NN 0.0 0.0 5.H 0.H 00.N 00.0 N mumuHaHomum 0.5H H.0N 0.00 0.00 5.NN 0.0N N5.N 50.0 N uomumohmmom 0.NN 0.N0 0.N0 0.00 0.0H 0.0N 00.0 00.0 H mumuHmHomum 0.5H H.0N 0.00 0.00 0.0N H.H0 00.N 00.0 H usmumcumasm N.0N 0.00 0.00H 0.00H 5.00 0.00 0N.0 N5.0 00H=0 0.0H 0.0N III III 0.00H 0.00H N0.N 00.0 00H00H< 0000 0000 soon 0000 soon 0000 soon 0000 z oHnmuumuuxm z HmonHuo maOHuomuw 06 no eANV mo 0H000u0100sn0 wwwucmoumm wwwuamoumm cowouqu :oHuaHuomma mMHMMHm 00 mGOHuomuuxm Hmu0>0m Baum 000Hmuno chHuomum 000 00 mUHumHumuomumno 000ouqu .H0 0H00H 129 Spencer et al. (1969), and Knuckles at al. (1970) for alfalfa juice. The nitrogen contents of the chloroplastic fractions obtained by Singh (1967) and Subba Rau et al. (1969) by heating lucerne juice, but to a slightly higher temperature, are comparable to our value obtained from "good" alfalfa. However, Singh (1967) found this fraction to con- tain 40% of the total precipitable protein; whereas our preparation con- tained nearly two-thirds of the precipitable protein. Leaf protein concentrate prepared from alfalfa by various extrac- tion methods has been reported to have a crude protein content of 49-68% (Smith, 1966; Singh, 1967; Spencer at al., 1969, 1971; Subba Rau et al., 1969; Poppe et al., 1970) which is close to values obtained for the acid precipitate extracted from alfalfa in 'poor" condition and considerably less (about 20%) than values for the acid precipitate from "good" alfalfa. The acid precipitate would be more comparable to the cyt0p1asmic fractions isolated by some investigators. The crude protein contents of cytoplasmic fractions obtained by Singh (1967) and Subba Rau et al. (1969) by heat precipitation are slightly less than the average crude protein content of acid precipitates extracted from "good" alfalfa. Singh (1967) also reported that his cytoplasmic fraction contained 45-50% of the precipitable protein. In contrast, our acid precipitate contained only about one-third of the precipitable protein. The literature generally makes little mention of the final liquor. Spencer at al. (1969, 1971) reported a protein content (ranging from 16- 29%) for this fraction comparable to that obtained for the final super- natant from "good" alfalfa; although Poppe et al. (1970) found the protein content of the liquor to be not higher than 17%. In conclusion, in agreement with Festenstein (1961) and Byers and Sturrock (1965), yields were noticeably dependent upon the age and physical 130 state of the leaf and were considerably reduced after the plant had devel- oped seeds. Yields from plants which were dry and yellow were lower than those which were tender and green. The lower yields appeared to be a combined result of less nitrogen in the leaf, reduced extractability, and less acid precipitable material. The percentage of alfalfa leaf nitrogen extractable into the juice varies with the mechanics of the extraction process. The values obtained by this small scale laboratory method for "good" alfalfa are comparable to those obtained by some specially designed leaf extractors, e.g., Hartman et a1. (1967), Singh (1967), and Knuckles at al. (1970). Although workers may often report the percentage of nitrogen extractable into the juice, they generally fail to give the percentage of nitrogen precipitable by heat or acid. One exception was Chayen at al. (1961) who reported that by their industrial impulse rendering process 22-64% of lucerne nitrogen is obtained in the protein precipitate. A sum- mation of the chlor0plastic and acid precipitates from the "poor" alfalfa would give a value equal to the lower end of the range reported by Chayen at al. (1961). The "good" alfalfa gave a value in the middle of that range. From one pound of wet alfalfa in good condition, 3.3 and 3.7 g of dried protein were obtained as precipitate 3 and as supernatant 3, respectively. Quality and Nutritive Value Chemical Evaluation Gross composition. The proximate composition of the acid precipi- tate and acid supernatant from each cutting are given in Table 32. All values were corrected for moisture in the freeze-dried samples. Protein was calculated as nitrogen times 6.25; 6.25 being the classical conversion factor for an ideal, non-conjugated, 16% nitrogen, protein. Unfortunately no 100% satisfactory method has been formulated for determining the 131 Table 32. Proximate compositions of alfalfa leaf protein fractions *7 Fraction 1P3 2P3 3P3 183 283 383 Protein 81.50 68.34 66.14 33.54 22.41 20.57 Carbohydrate 6.54 9.10 9.43 35.51 32.63 30.64 Lipid 1.08 1.66 2.44 0.95 3.11 3.23 Ash 1.24 1.46 1.52 5.20 6.10 4.67 * see methods section, p. 79, for explanation of nomenclature. total carbohydrate content of LPG (Byers, unpublished). Lipid includes a wide variety of compounds and consequently concentrations reported will depend upon the particular solvent system employed. Chloroform— methanol (2:1 v/v) was used, as many workers have found this solvent system to be satisfactory for removing most lipid materials from LPC (Byers, unpublished; Lea and Parr, 1961; Lima, Richardson, and Stahmann, 1965; Shah, 1968). The acid precipitate contained approximately 72% protein, 8% carbo- hydrate, 2% lipid, and 1.5% ash. The protein contents were compared with literature values for fractions extracted from alfalfa in the previous section on yields. In review, the protein contents are greater than the whole LPC preparations; and slightly greater than (in the case of the first cuttings) and slightly less than (in the case of the second and third cuttings) the cytoplasmic fractions isolated from alfalfa juice. Nonprotein constituents are not frequently reported. Smith (1966) reported a slightly greater value for carbohydrate (12.5%) in LPC obtained by the impulse rendering of alfalfa than that obtained for the acid precipitates. Subba Rau et a2. (1969) reported an even greater carbohydrate content (16%) for their cytoplasmic fraction. 132 However, this was obtained by difference rather than by analysis. Most preparations of LPC have high contents of.lipid--12-25% (Pirie, 1960; Lea and Parr, 1961; Smith, 1966; Shah, 1968; Subba Rau et al., 1969)-- because the concentrate, unless solvent extracted, contains chlorophyll and associated lipids. However, Spencer at al. (1969, 1971) reported the fat content of their protein-chlorophyll-xanthophyll concentrate (PRO-XAN) to be only 7-9%. Subba Rau at al. (1969) reported their cyto- plasmic fraction to be low in lipid (5%). However, this is still slightly greater than contents obtained for the acid precipitates; although the ash contents of the two are similar. On the other hand, the PRO—XAN concentrate was reported by Spencer at al. (1969, 1971) to have a much higher ash content (IO-13%). In contrast, the acid supernatant contained only about 25% protein, 33% carbohydrate, 2.5% lipid, and 5.5% ash. The only literature com- parison that can be made for the acid supernatant is with the alfalfa solubles obtained as a by-product of the wet fractionation procedure. Spencer et al. (1969, 1971) reported comparable protein contents. The supernatant contained a somewhat greater lipid content and about four times less ash than the alfalfa solubles. Both fractions showed changes in composition between cuttings (see Figures 18 and 19). Most striking was the decrease in protein content from the first to second cuttings; the acid precipitates drapped from 82-68% and the acid supernatants from 34-22%. The decrease was only about 2% between the second and third cutting. However, commensurate increases in carbohydrate, lipid, or ash contents were not encountered. There was a small increase in the carbohydrate content of the acid pre- cipitate from the first to third harvest. The carbohydrate content of the acid supernatant, on the other hand, gradually decreased from 36-31Z. 133 IOO - EBC)F' 4 protein 60- ‘7. 4o- 20r- carbohydrate 9 fl 0 l I 1 l l J | 2 3 l 2 3 CUTTING Figure 18. lipid contents of the acid precipitate from the first through the third cutting. Changes in protein, carbohydrate, ash, and 4o — — carbohydrate ash 30 " 70 20 " protein lipid I0 '- I 2 3 I 2 3 CUTTING ' Figure 19. Changes in protein, carbohydrate, ash, and lipid contents of the acid supernatant from the first through the third cutting. °/o 134 Both fractions exhibited increases in total lipid. Pirie (1961c) reported that lipid content is independent of the number of cuttings. The ash content was essentially constant for all regrowths. These four components accounted for from 80-90% of the precipitate fractions but for only from 60-75% of the supernatant. It is conjectured that the supernatant fraction contains significant amounts of phenolic and polycyclic type compounds, e.g., tannins, saponins, etc., which have been found in plant extracts (cf. Singleton and Kratzer, 1969; Birk, 1969). Amino acid composition. The amino acid compositions of the acid precipitate and acid supernatant from each cutting are given in Table 33 and are compared with the essential amino acid composition of whole egg protein. Composition expressed in males of amino acid per 1000 moles amino acids and in grams of amino acid per 100 grams of sample are given in Tables A2 and A3, respectively, of the appendix. The precipitate and supernatant, as might be expected, show some— what differentdistributions of amino acids. In general, the precipi- tates have greater contents of Met, Leu, Tyr, phenylalanine (Phe) and Arg, and lesser amounts of Cys, aspartic acid (Asp), serine (Ser) and proline (Pro) than the supernatants. Neither fraction exhibits unusually high or low concentrations of any particular amino acid residue. Amino acid compositions which have been published are generally of LPC which was prepared by direct precipitation from the juice; and therefore contains chloroplastic proteins. Nevertheless, the acid pre- cipitate shows a distribution of amino acids very similar to that reported by Gerloff et al. (1965) for alfalfa LPC obtained by impulse rendering, except that the Met contents are twice those given by Gerloff at al. (1965). This is understandable since they did not determine Met Table 33. 135 Amino acid composition of alfalfa leaf protein fractions expressed as grams of amino acid/100 grams protein Amino acid Protein fraction 193*: 2P3 3P3 133 253 333 Egg** Lysine 6.60 6.89 9.21 8.24 7.53 6.76 6.4 Threonine 5.17 5.14 5.17 6.08 6.10 5.74 5.1 Valine 6.94 7.17 6.43 6.41 5.49 5.65 7.3 Methionine 3.11 2.47 2.39 1.82 1.93 1.07 --- Cystine 1.15 1.19 1.12 3.37 3.05 3.75 —-- (met + cys) 4.26 3.66 3.51 3.64 3.86 2.14 5.5 Isoleucine 5.16 5.73 4.92 5.00 4.52 4.55 6.6 Leucine 8.78 9.10 8.35 6.61 5.63 5.74 8.8 Tyrosine 4.35 4.45 5.03 2.83 3.51 3.16 --- Phenylalanine 6.38 6.69 5.99 4.14 4.18 4.16 --- (phe + tyr) 10.73 11.14 11.02 6.97 7.69 7.32 10.0 Tryptophan 2.48 2.91 2.51 1.85 2.94 2.72 1.6 Histidine 3.31 2.68 3.01 2.19 2.43 2.17 --- Arginine 7.93 7.46 7.15 3.87 4.19 3.72 —-- Aspartic acid 9.92 10.46 9.86 12.93 13.76 14.91 -—- Serine 2.58 3.07 3.19 5.51 6.02 5.41 ——- Glutamic acid 11.33 11.65 10.63 11.97 12.51 13.73 -—- Proline 4.45 4.54 4.11 5.49 6.00 5.96 --- Glycine 4.72 4.87 4.44 5.71 5.27 5.60 --- Alanine 5.64 3.51 6.49 5.97 4.93 5.22 --- * see methods section, p. 79, for explanation of nomenclature. ** FAQ/WHO, Protein Requirements, 1965. 136 on samples oxidized prior to acid hydrolysis. The acid precipitate also shows a similar composition, excepting Met and Trp, to that reported by Smith (1966) for alfalfa cytoplasmic protein. Again this difference is probably due to differences in analytical techniques. However, the acid precipitate does not show such close agreement with the amino acid composition of lucerne cyt0p1asmic protein determined by Wilson and Tilley (1965). The acid precipitate had higher concentrations of Met and Phe and a much lower content of Arg than the cyt0p1asmic fraction. The protein liquor is generally discarded and the author is not aware of any published amino acid compositions of this fraction. Hart- man et al. (1967) published the amino acid composition of alfalfa juice extracted from two cuttings of alfalfa. The amino acid distribution for the juice is, with the exception of Met, similar to that obtained for the acid precipitate. There is a muCh larger variance noticeable between the juice and the acid supernatant; particularly with Lys and Cys in the first cutting, leucine in the second cutting, and Phe and Arg in both cuttings. There is little difference in the amino acid composition of similar fractions prepared from different cuttings. Although significant work is lacking on amino acid composition of alfalfa LPC extracted from regrowths, leaf protein generally shows a more or less constant amino acid distribution irrespective of species or maturity (cf. Gerloff et al., 1965; Pirie, 1966c). However, a few trends are apparent. The acid precipitates prepared from successive cuttings during the season showed an increase in Lys of nearly 40% with smaller increases in Ser (24%) and Tyr (16%) and decreases in Met (23%) and Arg (10%). The super- natant exhibited small increases in two nonessential amino acids--Glu (15%) and Asp (15%)--with a decrease in Lys (18%). The amino acid 137 distribution in alfalfa juice from the first and second cuttings also showed reasonable consistency (Hartman et al., 1967). The major change from the first to second cutting was a 30% increase in Cys; smaller variations were in Lys (+16%), Arg (+13%), Met (-18%), and Asp (-18%). The only correspondence between the present study and these literature values appears to be the increase in Lys and decrease in Met in the acid precipitate from the first to the second cuttings. To evaluate the potential nutritional value of a protein, a compari- son can be made of its essential amino acid pattern relative to a standard protein of known "high" nutritive value. The chemical score (CS) expresses the quality of a protein in relation to its limiting amino acid, i.e., that essential amino acid present in smallest amount in relation to its requirement, and thus, becomes a means of predicting maximum theoretical efficiency of protein utilization. This value is derived by computing the concentration ratios of the individual essential amino acids between the test protein and a reference protein (Mitchell and Block, 1946). The lowest ratio (x 100) is the chemical score. Unfortunately, no perfect reference protein exists (Cresta et al., 1969). In 1957 FAQ published a provisional essential amino acid pattern which has been used as a reference protein. The other most frequently used reference proteins are milk (either cow's or human) and egg. In 1965 FAG/WHO adopted the essential amino acid pattern of whole egg protein in preference to the 1957 provisional pattern. Cresta et al. (1969) concluded that when the CS is used to estimate the ability of a food protein to maintain the nitrogen balance the pattern of milk protein is more suitable than egg protein. However, when the CS is used as an estimation of the capacity of the proteins to meet growth requirements (protein efficiency ratio), the egg protein pattern appears to be more satisfactory. 138 Chemical scores for the acid precipitates and supernatants calcu- lated against all four reference proteins are listed in Table 34. For the purpose of these calculations, Cys and Tyr were included at a level not to exceed that provided by Met and Phe, respectively. Table 34. Chemical scores of alfalfa leaf protein fractions Reference protein* Leaf fraction 1P3 2P3 3P3 153 283 3S3 1957 FAO provisional pattern looa** 87a 84a 87a 92a 513 Egg 78a 67a 64a 66a 71a 39a Cow's milk 31b 88c 77b 67d 57d 58d Human milk 81b 85a 77b 69e 63d soa * essential amino acid patterns from FAD/WHO, Table 6 (1965) ** a b c d e limiting amino acid: S-aa; ile; lys; leu; aromatic aa Chemical scores calculated in reference to the provisional pattern and egg in all cases were based on the sulfur-amino acids as the limiting amino acid. Methionine has been stated to be the limiting amino acid in LPC by several investigators (cf. Gerloff et al., 1965; Singh, 1967; Sentheshanmuganathan and Durand, 1969). Against the provisional pattern, all fractions with the exception of 383 have relatively high CS, i.e., between 84 and 100. These values exceed those for cereal grains, oil- seeds, casein, and are approximately equal to beef liver (FAQ/WHO, 1965). With egg protein as a reference, the relative order of CS is the same but the absolute values are lower (64—78, excepting 3S3), but are still not sufficiently low to be classed as poor. Smith (1966) reported the chemical scores of alfalfa LPC and cytoplasmic protein to be 57 and 65, 139 respectively, with the sulfur-amino acids limiting. Since milk (cow and human) protein is also considered limiting in the sulfur-amino acids, CS calculated relative to milk are generally slightly higher than those based on egg. Thus, in 1P3 and 3P3 Ile becomes limiting and in 2P3 either Lys or the sulfur-amino acids are limiting. The supernatant fractions, when referred to cow's milk, show Leu to be limiting, and when referred to human milk, show the aromatic-amino acids, Leu and the sulfur-amino acids to be limiting for the first, second, and third cuttings, respectively. With a CS range of 77-88 against milk and 64-78 against egg, the acid precipitated protein might be better for maintenance than for growth according to the criteria of Cresta et al. (1969). However, CS values for the supernatant fraction relative to milk are within the range of egg, thus it would be considered as a protein source equivalent for maintenance and growth. Ranking of the six samples in order of decreas- ing protein quality as evaluated by CS values obviously depends on which protein is used as a reference. The CS is an adequate first approximation for screening protein for nutritional quality. However, the entire essential amino acid picture should be considered for more meaningful evaluation. The ratios (as percentages) of each essential amino acid for each fraction to each standard--FAO, egg, cow's milk, and human milk-—are given in Tables 35, 36, 37, and 38, respectively. Each of the non-limiting essential amino acids is present in an amount greater than the 1957 FAO provisional pattern for all fractions studied. This was also noted by Buchanan (l969b) for wheat LPC and by Sentheshanmuganathan and Durand (1969) for LPC extracted from various Indian plants. Relative to egg, Val and Ile represent the second and 140 Table 35. Percentage of each essential amino acid as compared to that in the 1957 FAO provisional pattern Amino acid Fraction 1P3 2P3 3P3 183 233 353 Lys 100 A; Thr 100 _g) Met + (Cys) 100 87.1 83.6 86.7 91.9 51:0 Val 100 -% Ile 100 ;% Leu 100 _; Phe + (Tyr) 100 g? Trp 100 _%> Table 36. Percentage of each essential amino acid as compared to that in egg Amino acid Fraction 1P3 2P3 3P3 183 283 353 Lys 100 ) Thr 100 _; Met + (Cys) 77.5 66.5 63.8 66.2 70.2 38.9 Val 95 96.8 88.1 87.8 75.2 77.4 Ile 78.2 86.8 74.5 75.8 68.5 68.9 Leu 100 100 94.9 75.1 64.0 65.2 Phe + (Tyr) 100 i’ 69.7 76.9 73.2 Trp 100 __g’ 141 Table 37. Percentage of each essential amino acid as compared to that in cow's milk Amino acid Fraction 1P3 2P3 3P3 183 233 353 Lys 85.4 88.3 100 100 96.5 86.7 Thr 100 gig’ Met + (Cys) 100 *4; 64.8 Val 100 100 93.2 92.9 79.6 81.9 Ile 80.6 89.5 76.9 78.1 70.6 71.1 Leu 88.7 91.9 84.3 66.8 56.9 58.0 Phe + (Tyr) 100 100 94.9 75.1 64.0 65.2 Trp 100 w Table 38. Percentage of each essential amino acid as compared to that in human milk Amino acid Fraction 1P3 2P3 3P3 183 233 3S3 Lys 100 ——1> Thr 100 i’ Met + (Cys) 99.1 85.1 81.6 84.7 89.8 49.8 Val 100 100 97.4 97.1 83.2 85.6 Ile 80.6 89.5 76.9 78.1 70.6 71.1 Leu 99.8 100 93.8 74.3 763.3 64.5 Phe + (Tyr) 100 a» 69.0 76.1 72.5 Trp 100 w 142 third, respectively, limiting essential amino acids in the acid precipi- tate. The other residues exist in concentrations comparable with egg. An analysis of the supernatant fraction shows that, in addition to the sulfur-amino acids, Val, Ile, Leu, and the aromatic residues are present in lower concentrations than in egg (64-88%). It is especially important to note that in both fractions threonine (Thr), Lys, and Trp, which are often limiting in real diets, particularly cereal, are in concentrations greater than that of egg. Therefore, these two fractions should be con— sidered as important potential supplements to diets low in these amino acids-e.g., corn meal, limiting in Trp; rye, limiting in Thr; oat, rice, limiting in Lys (FAD/WHO, 1965). It must be remembered that leaf protein would never be used as the sole source of nitrogen. Therefore, its ability to complement the amino acid composition of other common dietary proteins should also be considered. Chayen et al. (1961) stated that supplementation of other protein foodstuffs appears to be the greatest potential value of leaf protein preparations. Compared to milk, the concentrations of the non-limiting essential amino acids of the acid precipitate vary from 82-100% of those in the reference; whereas for the supernatant fraction, they range from 64-100% of those in the reference. The precipitate appears to have a more desirable amino acid distri— bution and possesses a lower ratio of non-essential to essential amino acids. The ratios (N/E) are as follows: 1P3 1.00 .133 1.23 2P3 0.93 283 1.29' 3P3 0.96 3S3 1.46 The total essential amino acid pattern is considered when calculat- ing the essential amino acid index (EAA-I) which is a more sophisticated type of chemical score. This value is defined as the geometric mean of 143 the ten essential amino acid (includes Arg and His) egg ratios; the egg ratio being the ratio of an essential amino acid in a protein relat- ive to its concentration in whole egg protein (Oser, 1951). EAA-I values for the leaf fractions are listed with some standard proteins calculated by Oser (1951): 1P3 94.6 1S3 81.1 Egg 100 2P3 94.4 283 80.4 Casein 89 3P3 91.2 383 74.2 Wheat 67 As was the case for the CS, the EAAeI is more valuable in distin- guishing wide variations than narrow ones, but gives the same general conclusion, i.e., "good" for the precipitate, "fair" for the supernatants from the first and second cuttings, and "poor" for the supernatant from the last cutting. The general conclusion reached in the literature is that LPC has a favorable balance of essential amino acids and non-essential amino acids-~except Met; therefore, it should be a well balanced source of dietary protein if supplemented with Met. It would also be a better supplement than most of the seed proteins but not as good as casein or egg (cf. Gerloff et al., 1965; Stahmann, l968a,b; Pirie, 1969b). A similar conclusion.was reached here for the acid precipitate except that Met did not appear to be as limiting; in fact, the Met con- tents were similar to that of milk. Since Smith (1966) has shown the amino acid compositions of alfalfa chloroplastic and cytoplasmic frac- tions to be similar, the slightly better nutritive value, as indicated by composition, for the acid precipitate is probably due to the identifi- cation of Met as Met sulfone rather than to the absence of the chloro- plastic proteins. Composition alone can be a misleading indicator of nutritive value. Chemical evaluations of protein quality are primarily based on the premise 144 that the pattern of amino acids absorbed during the process of digestion reflect the total amino acid composition of the ingested protein. Thus, they do not evaluate digestibility, absorption, and utilization of the protein or other components in the protein source. Biological Evaluation Protein efficiency ratio. The nutritive value of the fractions can be evaluated better by biological assays than by examining the amino acid composition alone; for bioassays provide data reflecting the sum total of the protein's utilization. One common method is measuring the rate of growth of an animal under defined conditions. If there is a deficiency of one or more essential amino acids, growth will be reduced (McLaughlan and Campbell, 1969). The particular assay chosen for this study was the protein efficiency ratio-PER (Osborne, Mendel, and Ferry, 1919). In this assay, growth is related to protein intake. Although enjoying widespread use, the PER assay elicits one major criticism; namely, that no allowance is made for maintenance. However, no matter what method of protein quality evaluation is used, the various proteins are placed into the same relative quality ranges in regard to protein nutrition (Bergen, 1971). Therefore, despite its limitations, the PER method appears to be as precise and simple as any protein evaluation presently available (McLaughlan and Campbell, 1969). The rat is the classical animal used for biological assays of pro- tein quality. However, in this study sufficient quantities of the leaf protein preparations were not available to conduct conventional feeding studies with rats. Therefore, the much smaller meadow vole (Micnotus pennsylvanicus) was chosen as the experimental animal. Only 5 g of diet 145 at a 7% level of protein per day are required and a PER can be obtained in five or six days. Elliott and associates (Elliott, 1963a; Markarian and Elliott, 1968; Shenk, Elliott, and Thomas, 1970) demonstrated that the meadow vole can be used successfully for bioassays. It has been used primarily in studies with forage plants (Elliott, 1963a; Schillinger and Elliott, 1966a). The PER for each diet obtained from replications l, 2, 4, 5, and 7 are reported in Table 39. The mean value and its standard error for each protein source are also included. Analysis of variance could not be done meaningfully due to insufficient replications and heterogeneous variance. However, five replications do give relatively reliable mean values. Replication 3 was not useable as the three animals on the acid precipitate fractions emptied their water bottles. The resulting dehy- dration was no doubt partially responsible for the Observed loss of weight or reduced gain. Protein efficiency ratios of replication 6 were somewhat erratic, including the control (PER of less than 1), indi- cating a poor growing litter. The control for replication 5 killed Table 39. Protein efficiency ratios of alfalfa leaf protein fractions and casein Replication Protein source Casein 1P3ll 2P3 3P3 183 283 383 Two 2.50 2.05 2.40 2.62 --- died -0.24 Four 2.72 1.79 2.82 3.18 --- died 0.27 Seven 3.55 3.43 2.93 3.37 1.61 died died Mean 2.49 2.07 2.46 2.33 1.87 died 0.19 Standard error 0.34 0.35 0.17 0.42 0.18 0.0 0.45 * see p. 79 for explanation of nomenclature. 146 itself on the fourth day by climbing up into the top of his feeder. Therefore, an average of the control values for replications l, 2, 3, 4, and 7 was used for this litter. There is no value for 183 in the fourth replication because the animal escaped. The differences in growth rates between litters (Whitmoyer, 1956) is immediately apparent, emphasizing the requirement for placing one animal of each litter on a different diet rather than one diet per litter; and for including a control for comparison in each litter. The casein PER mean of 2.49 agrees with the accepted PER value of 2.5 (Elliott, personal communication). This result gives validity to the PER means obtained for the experimental diets. Consider first the acid precipitates. The fraction prepared from the second cutting had a PER of 2.46 which is comparable to casein; and the PER from the third cutting fraction was 2.33 which is only slightly lower than that for casein. Thus these two fractions appear to possess protein of good biological quality. The acid precipitate obtained from the first cutting gave a lower PER (2.07) than casein or the same frac— tions from latter cuttings, but still was greater than 2.0.* Thus, the biological assays reflected the general conclusion reached by amino acid analyses, i.e., that the acid precipitate appears to be a protein with a nutritive value close to casein. Both are limiting to a similar degree in the sulfur-amino acids. In the literature most workers claim LPC, undamaged by inappro—‘ priate drying methods, to have a nutritive value as great as or better than the best seed proteins or fish meal but not as effective for the * PERs greater than 3, from 2-3, from 1-2, and less than 1 indi- cate protein of excellent, good, fair, and poor nutritive value, respectively. 147 nutrition of young animals as casein or egg (cf. Pirie, 1968; Henry and Ford, 1965). Shurpalekar et a1. (1967) reported freeze-dried lucerne LPC, unsupplemented with vitamins or minerals, to have a PER of only 1.54, significantly less than skim milk powder. This value was increased to 2.77 with methionine supplementation. Singh (1969) reported a PER range of only 1.3-1.8 for lucerne LPC, which is less than that obtained from the acid precipitates. Subba Rau et a2. (1969) reported 3 PER (1.36) for lucerne LPC in the range of Singh (1969). However, their cytoplasmic fraction appeared to have a higher nutritive value as evidenced by a PER of 2.02. This is still somewhat lower than values obtained for our comparable acid precipitate fractions. The reasons for the differences in PER between the first and last two cuttings were not obvious from amino acid composition. Variance in nutritive value, as evaluated by biological assay, between regrowths has not been systematically investigated. Byers (1967b) reported in her in vttro digestibility study that the results from regrowths were erratic. Both Woodham (1965) and Singh (1969) noted that different batches of protein preparations from the same species of plants may vary in nutritive quality. In this study the last two cuttings gave lower yields and precipitates with lower protein contents than the first cut- ting, yet yielded a protein of slightly higher nutritive value when evaluated biologically. This agrees with the statement of Byers (1961) that high extractability does not necessarily mean the product isolated is of high nutritive value and vice versa. Although the PER for leaf protein fed as the sole source of nitrogen to young animals is not necessarily a satisfactory estimate of nutritive value for humans, any bioassay is probably a better index of nutritive value for man than any in vitro method (Buchanan, l969b). However, the 148 value of a protein supplement, which leaf protein would be, depends not only on its limiting amino acid but on the excesses of essential amino acids in the supplement and how well this excess makes up deficiencies in the diet being supplemented. Therefore, although it was necessary and important to know the PER values of the protein fractions individually, their true worth as a dietary supplement can only be assessed when incor- porated into "practical" diets. Subrahmanyan and Sur (1949), Sur and Subrahmanyan (1954), Shurpalekar et al. (1967), and Doraiswamy et al. (1969) have shown that lucerne powder and lucerne LPC are good supple- ments to the poor Indian rice and ragi diets. Meaningful conclusions, based on only two replications, are diffi— cult to draw for the supernatant fraction obtained from the first cut- ting (133). Yet, these limited data indicate that this fraction is only fair in quality. Quite obviously, the animals did not prosper on the second cutting supernatant fraction (283); usually dying within 24-48 hr after initi- ation of the experimental diet. Their feed intake generally was some- what reduced (3-21 g) but proportional to the length of time they survived. Therefore, one can conclude that they did not die from star- vation due to rejection of the diet, but probably from a toxic component in the diet. This toxic factor appears to be present in a lower con- centration in the supernatant fraction from the last cutting (383). With the exception of the last litter, these animals appeared to be normal although they were not responding in growth (PER - 0.19). And, unlike the animals on 283, their intake of food was not reduced below normal; see Table 40. The acid precipitate values from replication 3 are not included in this table because absence of water would exert an effect on diet intake. Intake was considered normal for all experimental diets, except for 283. 149 Table 40. Feed intakes (in g/day) of voles Replication Protein source Casein 1P3 2P3 3P3 133 2S3 333 One 3.8 3.9 3.9 4.9 --- --- 4.5 Two 3.3 3.3 3.6 5.0 --- --- 3.8 Three 3.8 --- —-- --- --- 2.8a 4.5 Four 5.0 3.9 4.2 4.2 --- 2.6a 4.5 Five --- 4.1 4.8 3.8 3.3 2.4b 4.3 Six 3.3 3.9 5.1 5.1 4.0e 4.2e 4.8 Seven 4.3e 4.2e 4.3e 5.0d 5.9c 3.5b --- Mg§§_ 3.9 4.0 4.3 4.7 4.4 3.1 4.4 d length of replication: 8one day; btwo days; cthree days; four days; efive days; all others 6 days Toxicity. In early feeding experiments, Cowlishaw et al. (l956a,b) reported that lucerne LPC contained a water soluble toxic factor (which was claimed to be saponin); its effect could be counteracted by adding cholesterol to the diet or by washing the precipitate with hot water. They also noted (Cowlishaw et al., 1956b) that the dried whole juice had a low nutritive value. In contrast, Hartman et al. (1967) claimed that spray-dried alfalfa juice had a nutritive value equal to that of LPC. However, their evaluation was based on chemical assays and in vitro digestions. Later, Subba Rau et al. (1969) determined the PER of spray dried lucerne juice to be only 0.34. One of the eight animals on this diet died and one half of them lost weight; thus, their results were similar to those obtained in this study for 333. Also, in both cases differences in amino acid composition were not sufficient to account for the nutritional variation between the soluble extracts (whole juice or acid supernatant) and the precipitated fractions. 150 Aqueous extracts of alfalfa have been shown to contain saponins, to which a variety of diverse physiological effects have been attributed (cf. Scardavi and Elliott, 1967; Birk, 1969). Therefore, saponin toxicity is a possible answer to the poor performance observed with animals on the supernatant diets. In the semiquantitative analyses for saponin level of the vole diets, the acid precipitates showed negligible to zero red blood cell hemolysis, whereas the acid super- natants exhibited an abnormally high degree of red cell hemolysis, indicating high saponin level. Quantitating the degree of red cell hemolysis caused by the supernatant fraction diets was difficult due to their natural brown color. Although only low saponin alfalfa plants were used, the water soluble saponins may have concentrated in the supernatant fraction during the extraction process. Saponin levels in the plant are generally highest in the second cutting (Elliott, personal communication; Hanson et al., 1963) which may be the reason for the increased toxicity of that supernatant fraction. Also, since the latter two cuttings had protein contents of only two-thirds that of the first cutting, only two-thirds as much of the first cutting fraction was required to achieve equivalent protein levels in the diets. Con- sequently, these animals (133) could have received a lower concentra- tion of the toxin, which may account for the fair performance the voles showed on 133 as compared to the obvious toxicity of 233 and 333. Aut0psies were performed in a further attempt to elucidate the cause of the toxicity of these diets and to determine the extent of their pathological effects on the voles. With replications five (vole fed 233) and six (voles fed casein, 133, 233, 333, and 3P3) only gross abnormalities were looked for. The vole fed casein exhibited a full body cavity with no fluid, a full and firm intestinal tract, a cecum 151 12.5 cm long with a greenish-yellow color, and a visible liver. Voles carry out some digestion in the cecum, therefore its size is of import- ance. The two voles which had been fed the very toxic 233 showed an absence of body fat, edema of the mesentery and intestinal wall and red fluid in the body cavity (ascites). The intestinal tract contained only a small amount of mucus (plus in one animal some white flecks of unknown origin similar to mucus). The cecums were of small diameter, shorter than normal (6 and 10.5 cm), and not completely full-containing only some semi-solid, brown, fecal-like material. The lungs, heart, kidney, and liver showed normal color, position, and shape. However, the liver of one of the voles was slightly smaller than normal and hidden under the thoracic cage. These symptoms are indistinguishable from those of starvation. However, feed intakes had indicated the latter not to be the cause of death. The voles fed 133 and 333 exhibited a few of the above symptoms but to a much lesser degree. Both showed a small amount of ascites. Again lungs, heart, kidney, and liver were normal except that the vole fed 133 had a smaller liver. This vole's intestinal tract was filled with the previously mentioned semi-solid material but it had a slight yellowish color (more normal). The tract of the vole fed 333 was fuller and showed better tone. Both cecums were full, but they varied in length-13 and 10 cm for the voles fed 133 and 333, respectively. The vole which had been fed the diet based on the acid precipitate from the third cutting showed no abnormalities--i.e., the body cavity was full with no fluid, the liver was normal color and size, and the cecum was 12 cm long and had a yellowish-green color. Histopathologic examination of those voles of the last replication showed similar lesions in the voles fed 233 and 353- These 152 consisted of necrosis of the mucosa of the duodenum. The epithelial cells were desquamated with only a few cells at the base of the mucosa. There appeared to be an excess of mucus production in the intestinal tract. The vole fed 233 also had an unidentified precipitate in the tubules of the medulla in the kidney. However, there was no necrosis of tubular epithelium. No significant lesions were found in the liver, kidney, or brain. In summary the histopathological examination indi- cated the primary lesion in the voles fed the supernatant fractions from the last two cuttings to be enteritis. Most of the published physiological effects of alfalfa saponins are concerned with growth depression in Chicks and ruminant bloat (cf. Scardavi and Elliott, 1967; Birk, 1969). However, Hanson et a1. (1963) did note that saponins were reported to have an irritating effect on mucous membranes. Sollman (1957) reported that saponins are not absorbed from the intestine so that acute oral doses produce only local effects--marked increase of mucus and moderate increase of fluid; death resulting from inflammation of the alimentary canal. This con- forms to the autopsy results, and in addition to the other character- istics which saponins and the supernatant fractions exhibit--growth depression, bitter taste, non-dialyzable, foaming in aqueous solution, and hemolysis of red blood cells--suggest that saponins probably exerted an important role in the toxicity of the supernatant fractions. But before any significant property is unequivocally attributed to a saponin, it should be isolated in pure form and an investigation of its composi— tion, structure, and activity carried out (Birk, 1969). I There are also other possible toxic agents which must be considered. An important group is products resulting from the "browning reactions." The reactants--amine, sugar, and phenolic groups-—are present in the 153 supernatant. Browning can result in the following characteristics also possessed by the acid supernatant: brown discoloration, bitter flavor, tendency to foam, loss of nutritive value, and most important--possible toxic effects (Friedman and Shibko, 1969). Although the voles eating 233 usually became listless and shivered and often became hunch-backed, one (replicate 5) exhibited hyperexcitability and continuously ran in circles—~similar to symptoms exhibited by mice fed browning reaction reductones (Cutting et aZ., 1960; Ambrose et al., 1961). The supernatant fraction was observed to contain non-coaguable and non-dialyzable peptides and/or protein. Ramirez and Mitchell (1960) believed alfalfa trypsin inhibitor to have these characteristics. Additional compounds which have been extracted from plant materials, including alfalfa, and shown to be toxic include alkaloids, phenols, tannins, and other unknown factors (Keeler, 1969; Elliott, 1963a,b; Singleton and Kratzer, 1969; Schillinger and Elliott, l966b). Undoubtedly several of these factors could have contributed to the toxicity of the supernatant fractions. Before further speculation is made as to the use of this by-product as a feed supplement (cf. Pirie, l966c; Spencer et al., 1969) further testing with various species of animals and char— acterization and separation of the toxic component(s) will obviously be necessary. Accgptability Several investigators (of. Guha, 1960; Byers et aZ., 1965; Oke, l966a) have claimed that green LPC can be successfully incorporated into several foods. Others (cf. Smith, 1966; Kinsella, 1970) have claimed the green color and grassy flavor to be objectionable. Since nutritive value has little meaning if the product is objectionable for aesthetic 154 or other reasons, the protein fractions were incorporated into plain sugar cookies with no attempt to hide flavor or color. The mean values (hedonic scale) and statistical significance for each of the char- acteristics evaluated—-color, flavor, texture, and general acceptability-- are given in Tables 41 and 42. Table 41. Mean values and statistical significance for organoleptic variables for the acid precipitate Level of * alfalfa * * * General protein (%) Color Flavor Texture acceptability 0.0 7.23A 7.10A 6.653’A 7.00A 1.0 5.72B 5.75B 6.00b’B 5.85B 2.5 5.55B 4.45C 5.23“C 4.83C 5.0 4.60C 3.05D 4.43d'c 3.53D * means with same superscript form a statistically homogeneous group: lower case letter, P - .05; upper case letter, P = .01 The cookies made from the acid precipitate had a peppery appearance which varied with the amount of supplementation. In comparison to a plain white cookie, it was not as acceptable but was not considered objectionable. Texture was rated similarly to color. With increasing supplementation a more noticeable grainy or branny texture appeared. General acceptability appeared to be most closely related to the flavor rating, but was always slightly higher (0.1-0.5 point). Both varied about one unit between levels, i.e., control about 7 (like moderately), 1% supplementation about 6 (like slightly), 2.5% supplementation about 4.5-5 (neigher like nor dislike), and 5% supplementation about 3-3.5 (dislike moderately to slightly). For these two variables, each of the concentration levels was statistically significantly different from each 155 other. It was necessary to add 5% alfalfa protein (i.e., substitution of 20% of the flour) to induce unacceptability. The principal objection was usually referred to as a mild grassy flavor, the degree of which increased with greater amounts of supplementation. This is in agreement with Kamalanathan et al. (1969), who found that the color and texture were relatively minor problems when they incorporated LPC into various preparations. The leafy flavor, however, had an adverse influence on the preparations. Pirie (1959c) and Morrison and Pirie (1960) main- tain that LPC has little or no flavor of its own and that it blends well with moderate levels of spice or other flavoring agents. Table 42. Mean values and statistical significance for organoleptic variables for the acid supernatant Level of * alfalfa * * * General protein (%) Color Flavor Texture acceptability 0.0 7.20A 7.40A 7.25A 7.33A 1.0 5.45B 1.75B 5.20B 2.40B 2.5 4.60B 1.25B 3.30C 1.55” * means with same superscript form a statistically homogeneous group: lower case letter, P = .05; upper case letter, P = .01. The acid supernatant was unacceptable at both levels of fortifica- tion. However, the medium to dark brown color was not objectionable. The texture at the 1% level was acceptable, but became hard and unaccept- able at the 2.5% level. Flavor appeared to be the major factor in determining general acceptability. Comments indicated that the flavor at all levels of fortification was objectionable due to its extreme and sharp bitterness. In both panels, the judges did not agree absolutely 156 on the numerical ratings but almost always rated the various samples in the same relative order. In conclusion, the acid precipitate can be substituted for flour (up to a level of 10%) in a plain cookie without objection. The acceptable level would undoubtedly increase if incorporated into a food with a stronger flavor, e.g., chocolate, peanut butter, or highly spiced dishes. Kamalanathan et al. (1969) noted that the flavor could be successfully masked by spices, banana being particularly effective. Their conclusion.was that for better acceptability LPC should be incor- porated into highly spiced or flavored preparations. Also acceptability appears to depend on the particular cultural food habits or prejudices of the consumer (cf. Buchanan, 1968; Pirie, l969b,d). CONCLUSION A relatively simple method was used to isolate two protein fractions--one acid insoluble and the other acid soluble--from juice extracted from fresh alfalfa. In brief: the alfalfa was blended with water and pressed, the resulting juice heated to 48 C followed by centrifugation to remove the chloroplasts, the pH of the supernatant raised to 8.5 and the precipitate removed by centrifugation, and lastly, this second supernatant was acidified and the two final fractions sepa- rated by centrifugation. No significant proteolysis was noted in the juice under the conditions of the extraction. Incubation of the juice with cellulase, amylase, or pectinase did not increase the purity (i.e., protein content) of the acid precipitate, the fraction of major interest. Extractability of nitrogen was largely dependent on the condition of the alfalfa at the time of harvest; being highest if the plant was young (1/10 bloom), green and tender. (However, biological assays showed the protein from the plants which exhibited lower extractability to have the higher nutritive value.) With alfalfa in good condition, nearly 60% of the nitrogen of the leaf was extracted into the juice and of this extracted nitrogen about 27% appeared in the acid precipitate and 30% in the acid supernatant. On the other hand, for partially yellowed or mature alfalfa relatively high in dry matter extractability was only about 40%, with only 21% of the extracted nitrogen being precipitable by acid and 40% remaining in the acid supernatant. Yields increased with higher nitrogen content in the leaf; but this increase resulted 157 158 from the greater amounts of nitrogen in the final fractions, not neces- sarily from increased extractability. The acid precipitate contained approximately 72% protein, 8% carbohydrate, 2% lipid, and 1.5% ash on a dry basis. In contrast, the dry matter of the acid supernatant contained only about 25% protein, 33% carbohydrate, 2.5% lipid, and 5.5% ash. There were compositional changes between fractions obtained from the three cuttings made during the summer. Most noticeable was the decrease in protein contents of both the acid precipitates and acid supernatants from the first to second cuttings. However, there was little variation in amino acid composition between fractions from different cuttings. The acid precipitate had a slightly more favorable balance of essential amino acids than the acid supernatant; although methionine was the limiting amino acid in both fractions from all three cuttings. Resulting chemical scores, based on egg protein, with the exception of the acid supernatant from the third cutting, ranged from 64-78. Particularly significant is the fact that in all fractions threonine, lysine, and tryptophan--often limiting in many human diets—-are present in concentrations greater than that of egg. As evaluated by protein efficiency ratio, the nutritive value of the acid precipitate varied slightly with cutting (PER - 2.1, 2.5, 2.3 for the first, second, and third cutting fractions, reapectively), that of the second cutting fraction being equal to that of casein. However, the supernatant fractions appeared to be toxic, the degree of toxicity highest with the second cutting fraction and lowest with the first cut- ting fraction. Necropsy indicated the primary lesion to be inflammation of the intestinal tract. Chemically these diets were shown to contain high levels of saponins. This is a probable cause of the toxicity; although other unknown factors, e.g., browning reaction products, should be considered. 159 The acid precipitate is light colored and fairly bland, and there— fore was acceptable in cookies when substituted for 10% of the flour. The acid supernatant was unacceptable, not because of its brown color, but due to its extreme bitterness. In conclusion, a protein concentrate, which has a nutritive value similar to casein with excellent potential as a supplement to cereal diets and is acceptable organoleptically, has been extracted from alfalfa. The mother liquor, however, has been shown to be toxic to weanling meadow voles and unacceptable in human food. BIBLIOGRAPHY BIBLIOGRAPHY , Akeson, W. R. and M. A. Stahmann. 1964. A pepsin pancreatin digest index of protein quality evaluation. J. Nutr. 83: 257. 1965. Nutritive value of leaf protein concentrate, an in vitro digestion study. J. Agr. Food Chem. 13: 145. 1966. Leaf protein concentrates: a comparison of protein production per acre of forage with that from seed and animal crops. Econ. Bot. 20: 244. Akinrele, I. A. 1963. The manufacture and utilisation of leaf pro- tein. J. W. Afr. Sci. Ass. 8: 74. Altschul, A. M. 1966. Preface, p. vii. In R. F. Gould (ed.), Wbrld Protein Resources, Adv. in Chem. Series, No. 57. American Chemical Society, Washington, D.C. 1969. Food: protein for humans. Chem. Eng. News 47 (49): 68. Ambrose, A. M., D. J. Robbins and F. DeEds. 1961. Acute and subacute toxicity of amino-hexose-reductones. Proc. Soc. Exp. Biol. Med. 106: 656. Amerine, M. A., R. M. Pangborn and E. B. Roessler. 1965. Principles of'Sensory Evaluation of'Fbod. Academic Press, New York. p. 451. Anderson, E. V. 1971. The new priorities, food: preventing hunger and malnutrition. Chem. Eng. News 49 (l8): 19. Anon. 1953. Protein extraction from green crOps. World Crops 5: 62. 1959. Industrial production of vegetable protein. Chem. Ind. 1959: 1343. 1967. Green bread is good to eat. New Scientist 36: 225. 1969. Big squeeze: getting the most out of alfalfa. Agr. Res. 18 (3): 6. 1970. The challenge of synthetic and substitute foods, a symposium. Lebensmittel-Wissenschaft und Technologie (Food Sci. Technol.) 3 (2): II. 160 161 Arkcoll, D. B. and G. N. Festenstein. 1971. A preliminary study of the agronomic factors affecting the yield of extractable pro- tein. J. Sci. Food Agr. 22: 49. Armstrong, R. H. 1951. Amino acids in the proteins of herbage. J. Sci. Food Agr. 2: 166. Ayala, E. and E. L. Johnson. 1951. Vitamin B12 and growth inhibiting properties of the dried juice of alfalfa. Poul. Sci. 30: 893. Barber, R. 3., R. Braude and K. G. Mitchell. 1959. Leaf protein in rations of growing pigs. Proc. Nutr. Soc. 18: iii. Bergen, W. G. 1971. Human Nutr. Foods 927 Notes, Michigan State University, E. Lansing, Mich. Bickoff, E. M., A. Bevenue and K. T. Williams. 1947. Alfalfa has promising ohemurgic future. Chemurg. Dig. 6: 215. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.-74-76, USDA. p. 71. Bickoff, E. M., R. R. Spencer, 3. C. Witt, B. E. Knuckles, and J. B. Stark. 1968. Purine derivatives in alfalfa as growth stimu- lants for Bacillus subtilis. J. Agr. Food Chem. 16: 426. Birk, Y. 1969. Saponins, p. 169. In I. E. Liener (ed.) Toxic Con- stituents of'PZant Foodstuffs. Academic Press, New York. Borgstrom, G. 1967. The Hungry Planet. The Macmillan Co., New York. Borlaug, N. E. 1971. The future challenges in food production. Seminar, May 11, Mich. State Univ., E. Lansing, Mich. Brown, L. R. 1968. World food problems, p. 11. In R. I. Mateles and S. R. Tannenbaum (ed.) Single Cell Protein. MIT Press, Cambridge, Mass. Browning, J. E. 1970. Protein from forage crops? Chem. Eng. 77 (20): 100. Buchanan, R. A. 1968. The future of protein foods including leaf pro- tein. Austr. J. Dairy Technol. 23: 147. l969a. The production and potential of leaf protein. Food Technol. Austr. 21: 470. l969b. In vivo and in vitro methods of measuring nutrit- ive value of leaf protein preparations. Brit. J. Nutr. 23: 533. 1969c. Effect of storage and lipid extraction on the properties of leaf protein. J. Sci. Food Agr. 20: 359. Buchanan, R. A. and M. Byers. 1969. Interference by cyanide with the measurements of papain hydrolysis. J. Sci. Food Agr. 20: 364. 162 Byers, M. 1961. Extraction of protein from the leaves of some plants growing in Ghana. J. Sci. Food Agr. 12: 20. 19673. The in vitpo hydrolysis of leaf proteins. 1. The action of papain on protein extracted from the leaves of Zea mays. J. Sci. Food Agr. 18: 28. 1967b. The in vitro hydrolysis of leaf proteins. II. The action of papain on protein concentrates extracted from leaves of different Species. J. Sci. Food Agr. 18: 33. Unpublished. Practical details for the small-scale extraction of protein from leaves. Rothamst. Exp. Sta., Harpenden, Herts, U.K. p. 1-9. Byers, M., D. Fairclough and N. W. Pirie. 1955. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1955: 85. 1956a. The large-scale production of leaf protein. Biochem. J. 63: 33P. 1956b. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1956: 96. Byers, M., D. Fairclough, N. W. Pirie and W. E. Whitman. 1957. Large- scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1957: 101. Byers, M., S. H. Green and N. W. Pirie. 1965. Presentation of leaf protein on the table - II. Nutrition 19: 63. Byers, M., and J. W. Sturrock. 1965. The yields of leaf protein extracted by large scale processing of various crops. J. Sci. Food Agr. 16: 341. Carpenter, K. J. 1951. The relative nutritional values of animal and vegetable proteins for animals. Brit. J. Nutr. 5: 243. 1958. Chemical methods of evaluating protein quality. Proc. Nutr. Soc. 17: 91. Carpenter, K. J., J. Duckworth and G. M. Ellinger. 1952. The supple- mentary protein value of a by-product from grass processing. Brit. J. Nutr. 6: xii. 1954. Proc. Europ. Grassl. Conf. (Paris) 2: 243. Cited by Cowlishaw et al. (1956) in J. Sci. Food Agr. 7: 775. Carpenter, K. J. and G. M. Ellinger. 1955. Protein quality and "available lysine" in animal products. Poul. Sci. 34: 1451. Carpenter, K. J., G. M. Ellinger and D. H. Shrimpton. 1955. The evalu- ation of whaling by-products as feeding stuffs. J. Sci. Food Agr. 6: 296. 163 Central Food Technology Research Institute (CFTRI). 1968. Annual Report. Mysore, India. p. 21. Chakravorthy, P. R. and B. C. Guha. 1960. Proc. Symp. on Proteins (Mysore). p. 257. Cited by Valli Devi et al. (1965) in J. Sci. Food Agr. 16: 116. Chayen, I. H. and D. R. Ashworth. 1953. The application of impulse rendering to the animal-fat industry. J. Appl. Chem. 3: 529. Chayen, I. H., R. H. Smith, G. R. Tristam, D. Thirkell and T. Webb. 1961. The isolation of leaf components. I. J. Sci. Food Agr. 12: 502. Chibnall, A. C. 1922a. Investigations on the nitrogenous metabolism of the higher plants. Part II. The distribution of nitrogen in the leaves of the runner bean. Biochem. J. 16: 344. 1922b. Investigations on the nitrogenous metabolism of the higher plants. Part III. The effect of low temperature drying on the distribution of nitrogen in the runner bean. Biochem. J. 16: 599. 1922c. Investigations on the nitrogenous metabolism of the higher plants. Part IV. Distribution of nitrogen in the dead leaves of the runner bean. Biodhem. J. 16: 608. 1924. Spinacin, a new protein from spinach leaves. J. Biol. Chem. 61: 303. 1939. Protein Metabolism in the Plant. Yale Univ. Press, New Haven, Conn. Chibnall, A. C. and C. E. Grover. 1926. A chemical study of leaf cell cyt0p1asm. I. The soluble proteins. Biochem. J. 30: 108. Chibnall, A. C., E. J. Miller, D. H. Hall and R. G. Westall. 1933. The proteins of grasses. II. A new method of preparation. Biochem. J. 37: 1879. Chibnall, A. C. and L. S. Nolan. 1924. A protein from the leaves of the alfalfa plant. J. Biol. Chem. 63: 173. Chibnall, A. C., M. W. Rees and J. W. H. Lugg. 1963. The amino acid composition of leaf proteins. J. Sci. Food Agr. 14: 234. Chibnall, A. C. and S. B. Schuyver. 1921. Investigations on the nitrogenous metabolism of higher plants. Part I.‘ The isolation of proteins from leaves. Biochem. J. 15: 60. Cohen, M., W. Ginoza, R. W. Dorner, W. R. Hudson and S. Wildman. 1956. Solubility and colour characteristics of leaf proteins prepared in air or nitrogen. Science 124: 1081. 164 Conn, E. E. 1969. Cyanogenic glycosides. J. Agr. Food Chem. 17: 519. Coulson, C. B. and J. Davies. 1962. Saponins. II. Fractionation and pharmacological properties of lucerne saponins. The possible relation of these to bloat. J. Sci. Food Agr. 13: 53. Cowlishaw, S. J., D. E. Eyles, W. F. Raymond and J. M. A. Tilley. 1954. Nutritive value of lucerne protein concentrates, fed with and without cholesterol. Nature 174: 227. 1956a. Nutritive value of leaf protein concentrates. I. Effect of addition of cholesterol and amino acids. J. Sci. Food Agr. 7: 768. 1956b. Nutritive value of leaf protein concentrates. II. Effects of processing methods. J. Sci. Food Agr. 7: 775. Cresta, M., J. Périssé, M. Autret and E. Lombardo. 1969. Study of the correlations between biological and chemical measurements of food protein quality. Nutr. Newsletter (PAC) 7 (2): 1. Crook, E. M. 1946. The extraction of nitrogenous materials from green leaves. Biochem. J. 40: 197. Crook, E. M. and M. Holden. 1948. Some factors affecting the extrac- tion of nitrogenous materials from leaves of various species. Biochem. J. 42: 181. Crosby, D. G. 1969. Natural toxic background in the food of man and his animals. J. Agr. Food Chem. 17: 532. Cutting, W., A. Furst, D. Read, G. Read and H. Parkman. 1960. Circling syndrome produced in mice by dimehylaminohexose reductone. Proc. Soc. Exp. Biol. Med. 104: 381. Davies, M., W. C. Evans and W. H. Parr. 1952. Biological values and digestibilities of some grasses, and protein preparations from young and mature species, by the Thomas—Mitchell method, using rats. Biochem. J. 62: Proc. xxiii. Davys, M. N. G. and N. W. Pirie. 1960. Protein from leaves by bulk extraction. Engineering 190: 274. 1963. Batch production of protein from leaves. J. Agr. Eng. Res. 8: 70. 1965. A belt press for separating juices from fibrous pulps. J. Agr. Eng. Res. 10: 142. 1969. A laboratory-scale pulper for leafy plant material. Biotechnol. Bioeng. 11: 517. Davys, M. N. G., N. W. Pirie and G. Street. 1969. A laboratory-scale press for extracting juice from leaf pulp. Biotechnol. Bioeng. 11: 529. 165 Deijs, W. B. and J. J. I. Sprenger. 1952. (The production of protein from green plants.) Versl. Inst. Lanab. Onderz., Wageningen: 178. Cited by Raymond and Tilley (1956) in Colon. Plant Anim. Prod. 6: 3. Despande, P. D. and M. V. R. Rao. 1954. Nitrogen complex and amino acid composition of (i) amaranth (Amoranthus gangeticus) and (ii) aconite bean (Phaseolus aconitifblius). Ind. J. Med. Res. 42: 77. Doraiswamy, T. R., N. Singh and V. A. Daniel. 1969. Effects of sup- plementing ragi (Eleusine aoracana) diets with lysine or leaf protein on the growth and nitrogen metabolism of children. Brit. J. Nutr. 23: 737. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith. 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28: 350. Duckworth, J., W. R. Hepburn and A. A. Woodham. 1961. Leaf protein concentrates II: the value of a commercially dried product for newly-weaned pigs. J. Sci. Food Agr. 12: 16. Duckworth, J. and A. A. Woodham. 1961. Leaf protein concentrates 1: effect of source of raw material and method of drying on protein values for chicks and rats. J. Sci. Food Agr. 12: 5. Duckworth, J., A. A. Woodham and I. McDonald. 1961. The assessment of nutritive value in protein concentrates by the gross protein value method. J. Sci. Food Agr. 12: 407. Eggum, B. 0. 1970. The protein quality of cassava leaves. Brit. J. Nutr. 24: 761. Elliott, F. 0. 1963a. The meadow vole (Microtus pennsylvanicus) as a bioassay test organism for individual forage plants. Quart. Bull. Mich. Agr. Exp. Sta. 46: 58. 1963b. The isolation of anti—metabolites from individual alfalfa plants. 1. Cold water and paper chromatographic extraction techniques. Quart. Bull. Mich. Agr. Exp. Sta. 46: 242. Ereky, K. 1927. Process for the manufacture and preservation of green fodder pulp or other green plant pulp and of dry products made therefrom. Pat. (United Kingdom) No. 270,629. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.- 74-76, USDA. p. 71. Fairclough, D. and N. W. Pirie. 1954. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1954: 78. Ferguson, W. S. and E. R. Armitage. 1944. The chemical composition of bracken (Pteridium aquilinum). J. Agr. Sci. 34: 165. Festenstein, G. N. 1961. Extraction of proteins from green leaves. J. Sci. Food Agr. 12: 305. 166 Floyd, B. 1970. The 'green revolution': reality or myth. Cajanus 3: 205. Folch, J., M. Lees and G. H. Sloane-Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226: 497. FAQ. 1957. Protein requirements. FAO Nutritional Studies No. 16. FAQ, Rome. p. 21. 1964. Protein nutrition: needs and prospects, p. 98. In The State of'Fbod and Agriculture. FAO, Rome. FAO/WHO. 1965. Protein requirements. WHO Tech. Rep. Ser. No. 301. WHO , Geneva . Ford, J. E. 1960. A microbiological method for assessing the nutri- tional value of proteins. Brit. J. Nutr. 14: 485. 1962. A microbiological method for assessing the nutri- tional value of proteins. 2. The measurement of 'available' methionine, leucine, isoleucine, arginine, histidine, tryptophan and valine. Brit. J. Nutr. 16: 409. 1964. A microbiological method for assessing the nutri- tional value of proteins. 3. Further studies on the measurement of available amino acids. Brit. J. Nutr. 18: 449. Fowden, L., D. Lewis and H. Tristram. 1967. Toxic amino acids: their action as antimetabolites. Advan. Enzymol. 29: 89. Fox, H. C. and J. C. Waterlow. 1967. Nutritive value of leaf protein concentrates for children, p. 173. In H. D. Cremer (ed.) Nutrition under Various Geographic and Climatic Conditions, Vol 3 of Proc. Seventh Int. Congr. Nutr., Hamburg, 1966. Pergamon Press, New York. Freeman, 0. L. 1968. WOrld Without Hunger. Frederick A. Praeger, Inc., New York. Friedman, L. and S. I. Shibko. 1969. Adventitious toxic factors in processed foods, p. 349. In I. E. Liener (ed.) Tbxic Constituents of’Plant FOodstufjs. Academic Press, New York. Gerloff, E. D., I. H. Lima and M. A. Stahmann. 1965. Amino acid compo- sition of leaf protein concentrates. J. Agr. Food Chem. 13: 139. Gestetner, B., S. Shany, Y. Tencer, Y. Birk and A. Bondi. 1970. Lucerne saponins. II. Purification and fractionation of saponins from lucerne tops and roots and characterisation of the isolated fractions. J. Sci. Food Agr. 21: 502. Ghosh, J. J. 1967. Trans. Bose Res. Inst. 30: 215. Cited by Singh (1969) in J. Food Sci. Technol. 6: 165. 167 Glick, Z. and M. A. Joslyn. 1970. Food intake depression and other metabolic effects of tannic acid in the rat. J. Nutr. 100: 509. Gonzales, 0. N., L. B. Dimaunahan and E. A. Banzon. 1968. Extraction of protein from the leaves of some local plants. Philip. J. Sci. 97: 17. Goodall, O. 1936. Improvements relating to treatment of grass. Pat. (United Kingdom) No. 457,789. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc.,Agr. Res. Serv.-74-76, USDA. p. 71. Guggolz, J., V. V. Herring and G. 0. Kohler. 1967. Dehydrated alfalfa composition. Correlation of nonnutrient components with pro- tein content. J. Agr. Food Chem. 15: 1052. Guha, B. 1960. Leaf protein as human food. Lancet, 1.: 704. Hanson, C. H., G. O. Kohler, J. W. Dudley, E. L. Sorensen, G. R. Van- Atta, K. W. Taylor, M. W. Pederson, H. L. Carnahan, C. P. Wilsie, W. R. Kehr, C. C. Lowe, E. H. Stanford and J. A. Yungen. 1963. Saponin content of alfalfa as related to location, cutting, variety, and other variables. CrOps Res., Agr. Res. Serv.-34- 44, USDA. p. 1. Hanson, C. H., G. M. LOper, G. O. Kohler, E. M. Bickoff, K. W. Taylor, W. R. Kehr, E. H. Stanford, J. W. Dudley, M. W. Pederson, E. L. Sorenson, H. L. Carnahan and C. P. Wilsie. 1965. Variation in coumestrol content of alfalfa as related to location, variety, cutting, year, stage of growth, disease. USDA Tech. Bull. No. 1333. Cited by Ogden and Kehr (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.-74-76, USDA. p. 23. Hartman, G. H., Jr., W. R. Akeson and M. A. Stahmann. 1967. Leaf pro- tein concentrate prepared by spray drying. J. Agr. Food Chem. 15: 74. Heftmann, E. 1967. Biochemistry of steroidal saponins and glyco— alkaloids. Lloydia 30: 209. Henry, K. M. and J. E. Ford. 1965. The nutritive value of leaf pro— tein concentrates determined in biological tests with rats and by microbiological methods. J. Sci. Food Agr. 16: 425. Hits, C. H. W., W. H. Stein and S. Mbore. 1954. The amino acid compo- sition of ribonuclease. J. Biol. Chem. 211: 941. Hughes, G. P. and D. E. Eyles. 1953a. Extracted herbage leaf protein for poultry feeding. I. Introduction and feeding trials with laying hens. J. Agr. Sci. 43: 136. 1953b. Extracted herbage leaf protein for poultry feed- ing. 11. The use of leaf protein in chick rations. J. Agr. Sci. 43: 144. 168 Hussain, A., M. Ullah and B. Ahmad. 1968. Studies on the potentials of leaf proteins for the preparation of concentrates from various leafdwastes in West Pakistan. W. Pakis. J. Agr. Res. 6 (3): 110. Hylin, J. W. 1969. Toxic peptides and amino acids in foods and feeds. J. Agr. Food Chem. 17: 492. International Symposium of Protein Food and Concentrates (ISPFC). 1967. Recommendations. Central Food Technol. Res. Inst., Mysore, India. Isajev, F. 1956. (Testing of usefulness of protein-vitamin pastes in fodder.) Ann. Esl. Acad. Agric. Sci., Pl. Prod. Ser. 29: 721. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.-74-76, USDA. p. 71. Jackson, H. D. and R. A. Shaw. 1959. Chemical and biological proper- ties of a respiratory inhibitor from alfalfa saponins. Arch. Biochem. BiOphys. 84: 411. Jennings, A. C., A. Pusztai, R. L. M. Synge and W. B. Watt. 1968. Fractionation of plant material. 111. Two schemes for chemical fractionation of fresh leaves, having special applicability for isolation of the bulk protein. J. Sci. Food Agr. 19: 203. Jennings, D. M. and 0. A. M. Lewis. 1969. Methionine loss during pro— tein hydrolysis of plant material. J. Agr. Food Chem. 17: 668. Jennings, H. W. K. 1949. Improvements relating to methods of improving the taste of water soluble constituents of forage crops. Brit. Pat. 628,108. Cited by Tilley and Raymond (1957) in Herbage Abstr. 27: 236. Jones, M. 1969. Evaluation of alfalfa plants for saponin. Ph.D. dissertation. Mich. State Univ. E. Lansing, Mich. Kamalanathan, G., M. S. Usha and R. P. Devadas. 1969. Evaluation of the acceptability of some recipes with leaf protein concentrates. J. Nutr. Diet. 6: 12. Ke, B. and K. A. Clendenning. 1956. Properties of chloroplast dispersions in the presence of detergents. Biochim. Biophys. Acta.19: 74. Keeler, R. F. 1969. Toxic and teratogenic alkaloids of western range plants. J. Agr. Food Chem. 17: 473. Kelley, E. G. and R. R. Baum. 1953. Protein amino acids contents of vegetable leaf proteins. J. Agr. Food Chem. 2: 680. Kemble, A. R. and H. T. MacPherson. 1954. Mbnoamino monocarboxylic acid content of preparations of herbage protein. Biochem. J. 58: 44. Kinsella, J. E. 1970. Evaluation of plant leaf protein as a source of food protein. Chem. Ind. 1970: 550. Kirby, K. S. 1960. Induction of tumours by tannin extracts. Brit. J. Cancer 14: 147. 169 Kohler, G. 0., E. M. Bickoff, R. R. Spencer, 3. C. Witt and B. E. Knuckles. 1968. Wet processing of alfalfa for animal feed products, p. 71. In Tenth Tech. Alfalfa Conf. Procg,Agr. Res. Serv.-74-76, USDA. Kohler, G. O. and W. R. Graham. 1951. A chick growth factor found in leafy green vegetation. Poul. Sci. 30: 484. Knuckles, B. E., R. R. Spencer, M. E. Lazer, E. M. Bickoff and G. 0. Kohler. 1970. PRO-XAN process: incorporation and evaluation of sugar cane rolls in wet fractionation of alfalfa. J. Agr. Food Chem. 18: 1086. Krehl, W. A. and J. J. Barboriak. 1960. Factors affecting the utiliza- tion of food nutrients by the normal organism, p. 521. In R. S. Harris and H. V. Loesecke (ed.) Nutritional Evaluation Of'EOOd Processing. Wiley, New York. Kuppuswamy, 3., M. Srinivasan and V. Subramaniam. 1958. Rep. Ind. Counc. Med. Res., No. 33. p. 3221. Cited by Sentheshanmuganathan and Durand (1969) in J. Sci. Food Agr. 20: 603. Lea, C. H. and L. J. Parr. 1961. Some observations on the oxidative deterioration of the lipids of crude leaf protein. J. Sci. Food Agr. 12: 785. LeClerg, E. L. 1957. Mean separation by the functional analysis of variance and multiple comparisons. Agr. Res. Serv.-20-3, USDA. p. 23. Lendahl, I. L., W. T. Shalkop, R. W. Dougherty, C. R. Thompson, C. R. VanAtta, E. M. Bickoff, E. D. Walter, A. G. Livingston, J. Goggolz, R. H. Wilson, M. B. Sideman and F. DeEds. 1957. Alfalfa saponins. Studies on their chemical, pharmacological, and physiological prOperties in relation to ruminant bloat. USDA Tech. Bull. 1161. Cited by Birk (1969) in I. E. Liener (ed.) Toxic Constituents of’PZant Foodstuffs. Academic Press, New York. p. 180. Lepkovsky, 3., W. Shaeleff, D. Peterson and R. Perry. 1950. Alfalfa inhibitor in chick rations. Poul. Sci. 39: 208. Lewis, 0. A. M. 1966. Short ion—exchange column method for the estima- tion of cystine and methionine. Nature 209: 1239. Liener, I. E. 1958. Effect of heat on plant proteins, p. 79. In A. M. Altschul (ed.) Processed Plant Protein Foodstuffs. Academic Press, New York. ~ 1962. Toxic factors in edible legumes and their elimina- tion. Amer. J. Clin. Nutr. 11: 281. 1966. Toxic substances associated with seed proteins, p. 178. In R. F. Gould (ed.) World Protein Resources, Adv. Chem. Ser. No. 57. Amer. Chem. Soc., Washington, D.C. 170 1969. Introduction, p. 1. In I. E. Liener (ed.) Toxic Constituents of'Plant Foodstuffs. Academic Press, New York. Liener, I. E. and M. Kakade. 1969. Protease inhibitors, p. 7. In I. E. Liener (ed.) Toxic Constituents of'Plant Foodstuffs. Academic Press, New York. Lima, 1. H., T. Richardson and M. A. Stahmann. 1965. Fatty acids in some leaf protein concentrates. J. Agr. Food Chem. 13: 143. Loomis, W. D. and J. Battaile. 1966. Plant phenolic compounds and the isolation of plant enzymes. Phytochemistry 5: 423. Lugg, J. W. H. 1939. The representativeness of extracted samples and the efficiency of extraction of protein from the fresh leaves of plants; and some partial analyses of the whole proteins of leaves. Biochem. J. 33: 110. 1949. Plant proteins. Adv. Prot. Chem. 5: 229. Lugg, J. W. H. and R. A. Weller. 1944. Large-scale extraction of pro- tein samples reasonably representative of the whole proteins in the leaves of some plants: the amide, tyrosine, tryptophan, cystine (plus cysteine) and methionine contents of the prepara- tions. Biochem. J. 38: 408. 1948a. Partial amino acid compositions of some plant— 1eaf protein preparations: the arginine, histidine, and lysine contents. Biochem. J. 42: 408. 1948b. Protein in senescent leaves of Trifblium subter- raneum: partial amino acid composition. Biochem. J. 42: 412. Mahadeviah, S. and N. Singh. 1968. Leaf protein from the green t0ps of Cichorium intybus L. (chicory). Ind. J. Exp. Biol. 6: 193. Markarian, D. and F. C. Elliott. 1968. A bioassay of inbred pea lines for differences in nutritional value. Quart. Bull. Mich. Agr. Exp. Sta. 50: 293. McArthur, J. M. and J. E. Miltimore. 1964. Extraction of alfalfa leaf cyt0p1asmic protein with modified Pirie disintegrator. Can. J. Plant Sci. 44: 112. McDonald, A. N. C. 1954. Some experiments on the mechanical extraction of leaf protein. Rep. Nat. Inst. Agr. Eng. No. 43: 30. Cited by Tilley and Raymond (1957) in Herbage Abstr. 27: 236. McLaughlan, J. M. and J. A. Campbell. 1969. Methodology of protein evaluation, p. 391. In H. N. Munro (ed.) Mammalian Protein Metabolism, Vol. III. Academic Press, New York. Mickelsen, O. and M. C. Yang. 1966. Naturally occuring toxicants in food. Federation Proc. 25: 104. 171 1968. Naturally occurring toxic foods, p. 490. In M. G. Wohl and R. S. Goodhart (ed.) Modern Nutrition in Health and Disease. Lea & Febiger, Philadelphia. Miller, H. 1952. Method of processing alfalfa. U.S. Pat. 2,600,903. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. sens-74-76, USDA. p. 710 MitChell, H. H. 1924. A method of determining the biological value of protein. J. Biol. Chem. 58: 873. Mitchell, H. H. and R. J. Block. 1946. Some relationships between the amino acid contents of proteins and their nutritive value for the rat. J. Biol. Chem. 163: 599. Montgomery, R. D. 1969. Cyanogens, p. 143. In I. E. Liener (ed.) Toxic Constituents of'Plant Foodstuffs. Academic Press, New York. Mooijman, J. G. J. M. 1964. Purification and characterization of the trypsin inhibitor in alfalfa. Ph.D. Thesis, Cornell Univ. University Microfilms, Ann Arbor, Midh. Moore, 3., D. H. Spackman and W. H. Stein. 1958. Chromatography of amino acids on sulfonated polystyrene resins. Anal. Chem. 30: 1185. Moore, 3. and W. H. Stein. 1954. Procedures for the chromatographic determination of amino acids on four per cent cross linked sulfonated polystyrene resins. J. Biol. Chem. 211: 893. 1963. Chromatographic determination of amino acids by use of automatic recording equipment. Methods Enzymol. 6: 819. Morrison, J. E. and N. W. Pirie. 1960. Presentation of leaf protein on the table. Nutrition 14: 7. 1961. Large—scale production of protein from leaf extracts. J. Sci. Food Agr. 12: 1. 1963. The quick freezing and drying of moist protein, p. 132. In S. Cotson and D. B. Smith (ed.) Freeze Drying of’FbodL stuffs, a symposium. Columbine Press, Manchester, U.K. Cited by Pirie (1969) in Plant Foods Human Nutr. 1: 237. Munro, H. W. 1968. Nature of protein needs, p. 27. In R. I. Mateles and S. R. Tannenbaum (ed.) Single Cell Protein. MIT Press, Cambridge, Mass. National Academy of Sciences-National Research Council (NAS-NRC). 1966. Toxicants Occurring Naturally in Foods. Publ. No. 1354, Washington, D.C. Nazir, M. and F. H. Shah. 1966. Extractability of proteins from various leaves. Pakis. J. Sci. Ind. Res. 9: 235. 172 Nebraska Chemurgy Project. 1947. Fifth Annu. Rep., Univ. Nebr. p. 19. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.-74-76, USDA. p. 71. Oelshlegel, F. J., Jr., J. R. Schroeder and M. A. Stahmann. 1969a. Potential for protein concentrates from alfalfa and waste green plant material. J. Agr. Food Chem. 17: 791. l969b. Protein concentrates: use of residues as silage. J. Agr. Food Chem. 17: 796. Ogden, R. L. and W. R. Kehr. 1968. Field management for dehydration and hay production, p. 23. In Tenth Tech. Alfalfa Conf. Proc., Agr. RES. Sena-74-76, USDA. Oke, O. L. l966a. The introduction of leaf protein into the Nigerian diet. Nutrition 20: 18. l966b. Chemical studies on more commonly used leaf vege- tables in Nigeria. J. W. Afr. Sci. Ass. 11: 42. 1967. World Rev. Nutr. Diet. 8: 25. Cited by Oke (1969) in J. Nutr. Diet. 6: 37. 1968a. Protein supplement from green leaves. Food Technol. 22: 1097. 1968b. Unpublished data. Cited by Oke (1969) in J. Nutr. Diet. 6: 37. 1969. Leaf protein as a solution to protein shortage in developing countries. J. Nutr. Diet. 6: 37. Oke, O. L. and A. Tella. 1967. W. Afr. Pharm. (in press). Cited by Oke (1969) in J. Nutr. Diet. 6: 37. Onrust, H. 1969. The acceptability of non-conventional proteins. Voeding 30: 767. In Food Sci. Technol. Abstr. 2: 346. (1970) Osborne, T. B., L. B. Mendel and E. L. Ferry. 1919. A method of expressing numerically the growth-promoting value of proteins. J. Biol. Chem. 37: 223. Osborne, T. B., A. J. Wakeman and C. S. Leavenworth. 1921. The pro- teins of the alfalfa plant. J. Biol. Chem. 49: 63. Oser, B. L. 1951. Method for integrating essential amino acid content in the nutritional evaluation of protein. J. Amer. Diet. Ass. 27: 396. Pal, P. R. and B. c. Guha. 1953. sti. Cult. 18: 597. Cited by Valli Devi et all (1965) in J. Sci. Food Agr. 16: 116. Parini, V. P. 1953. Nauka i Zhizn', No. 11: 35. Cited by Raymond and Tilley (1956) in Colon. Plant Anim. Prod. 6: 3. 173 Parpia, H. A. B. 1967. Protein foods for less deve10ped countries, p. 53. In D. J. Tilgner and A. Borys (ed.) Proc. Second Int. Congr. Food Sci. Technol., 1966. Warsaw, Poland. Pederson, M. W., D. E. Zimmer, J. 0. Anderson and C. F. McGuire. 1966. A comparison of saponins from DuPoits, Lahontan, Ranger, and Unita, alfalfas. Proc. Tenth Int. Grassland Congr., Helsinki, p. 266. Cited by Birk (1969) in I. E. Liener (ed.) Toxic constituents of’PZant FOodstufTs. Academic Press, New York. Perry, T. W., R. C. Peterson and W. M. Beeson. 1969. The value of supplemental distillers solubles, fish solubles or alfalfa solubles in liquid supplements for fattening beef cattle. Rep. Annu. Ind. Cattle Feeders Day (March 28, 1969): 29. Cited by Spencer et al. (1969) at Amer. Soc. Agr. Eng. Meeting, Lafayette, Ind. (June 25, 1969). Peterson, D. W. 1950. Some properties of a factor in alfalfa meal causing depression of growth in chicks. J. Biol. Chem. 183: 647. Pirie, N. W. 1942a. The direct use of leaf protein in human nutri- tion. Chem. Ind. 61: 45. 1942b. Green leaves as a source of proteins and other nutrients. Nature 149: 251. 1952. Large-scale production of edible protein from fresh leaves. Rep. Rothamst. Exp. Sta. for 1952: 173. 1953. Food and the future. Part 3(A). The efficient use of sunlight for food production. Chem. Ind. 1953: 442. 1956a. A disintegrator for use with small quantities of fresh leaves. J. Agr. Eng. Res. 1: 81. 1956b. Unexploited technological possibilities of making food for man and animals. Proc. Nutr. Soc. 15: 154. 1957a. Biochemical engineering. Research 10: 29. 1957b. Leaf protein as a human food. Food Manuf. 32: 416. Reprinted in Nutrition 12: 17. (1958) 1958. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1958: 93. 1959a. Edible protein from leaves. Brit. Vegetarian 1(6): n.p. ' . 1959b. The large-scale separation of fluids from fibrous pulps. J. Biochem. Microbiol. Technol. Eng. (now Biotechnol. Bioeng.) 1: 13. 174 1959c. Leaf protein as human food. Lancet, II: 961. l959d. The extraction of protein from leaves. Rep. Rothamst. Exp. Sta. for 1959: 87. 1960. The extraction of protein from leaves. Rep. Rothamst. Exp. Sta. for 1960: 107. 1961a. Progress in biochemical engineering broadens our choice of cr0p plants. Econ. Bot. 15: 302. 1961b. The disintegration of soft tissues in the absence of air. J. Agr. Eng. Res. 6: 142. 1961c. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1961: 98. 1962. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1962: 102. 1963. Large-scale production of leaf protein. Rep. Rothamst. Exp. Sta. for 1963: 95. 1964a. Freeze-drying or drying by sublimation, p. 189. In D. w. Newman (ed.) Instrumental Methods of'Experimental Biology. Macmillan, New York. 1964b. Production of leaf protein as a human food. Rep. Rothamst. Exp. Sta. for 1964: 120. 1965. Production of leaf protein as human food. Rep. Rothamst. Exp. Sta. for 1965: 106. 1966a. Leaf protein as human food. Science 152: 1701. 1966b. Some suggestions on initiating work on leaf pro- tein as a human food. Rothamst. Exp. Sta., Harpenden, Herts., U.K. l966c. Leaf protein. International Biological Program (IBP) working group meeting, Warsaw, Poland, August 26-27, 1966. 1966d. The merits of food proteins from novel sources. Sci. Prog., Oxf. 54: 401. 1966e. Extraction of leaf protein. Rep. Rothamst. Exp. St. for 1966: p. 103. 19673. Leaf protein research in the International Bio— logical Program. Agr. Sci. Rev. 5 (4): l7. 1967b. Orthodox and unorthodox methods of meeting world food needs. Sci. Amer. 216 (2): 27. 175 1967c. Leaf protein and its by-products. Rep. Rothamst. Exp. Sta. for 1967: 115. 1968. Use of plant protein concentrates as human food. Chem. Ind. 1968: 864. l969a. Potential protein sources for human food. Abstr. Papers, Amer. Chem. Soc. 158: AGFD 1. l969b. The production and use of leaf protein. Proc. Nutr. Soc. 28: 85. .1969c. Complementary ways of meeting the world's protein need. Proc. Nutr. Soc. 28: 255. 1969d. The present position of research on the use of leaf protein as human food. Plant Foods Human Nutr. 1: 237. l969e. Large-scale protein preparations. Rep. Rothamst. Exp. Sta. for 1969 (Part I): 132. 1970. Thirty years progress with leaf protein. Cajanus 3: 279. Pleshkov, B. P. and L. Fowden. 1959. Amino acid composition of the pro- teins of barley leaves in relation to the mineral nutrition and age of plants. Nature 183: 1445. Poppe, J., P. P. Tabback and E. Maes. 1970. Factors influencing pro- tein extraction from lucerne (Medicago sativa). Lebensmittel- Wissenschaft und Technologie (Food Sci. TeChnol.) 3: 67. President's Science Advisory Committee (PSAC). 1968. The Wbrld FbOd Problem, Vol. II: Report of’the Panel on the Wbrld Food Supply. Govnt. Print. Office, Washington, D.C. Pudelkiewicz, W. J. and L. D. Matterson. 1960. A fat soluble material in alfalfa that reduces the biological availability of tocopherol. J. Nutr. 71: 143. Ramirez, J. S. and H. L. Mitchell. 1960. The trypsin inhibitor of alfalfa. J. Agr. Food Chem. 8: 393. Raymond, W. F. and J. M. A. Tilley. 1956. The extraction of protein concentrates from leaves. Colon. Plant Anim. Prod. 6: 3. Reber, E. F. and R. McVicar. 1953. The nitrogen composition of cereal grasses. III. Amino acid distribution in field clippings and growing plants. Agron. J. 45: 17. In Chem. Abstr. 47: 8189. Rouelle, H. M. 1773. J. Med. Chir. Pharm. 39: 262 and 40: 59. Cited by Pirie (1966) in Science 152: 1701. Russo, J. R. 1969a. Can new protein sources avert world shortage? Part 1. Food Eng. 41 (5): 74. 176 1969b. Can new protein sources avert world shortage? Part II. Food Eng. 41 (6): 80. Scardavi, A. and F. C. E-liott. 1967. A review of saponins in alfalfa and their bioassay utilizing Trichoderma sp. Quart. Bull. Mich. Agr. Exp. Sta. 50: 163. Schillinger, J. A., Jr. and F. C. Elliott. l966a. Bioassays for nutrit- ive value of individual alfalfa plants. Quart. Bull. Mich. Agr. Exp. Sta. 48: 580. 1966b. The isolation of anti-metabolites from individual alfalfa plants. III. An amino acid responding principle from Medicago falcata L. Quart. Bull. Mich. Agr. Exp. Sta. 48: 570. Schram, E., S. Moore and E. J. Bigwood. 1954. Chromatographic determina- tion of cystine as cysteic acid. Biochem. J. 57: 33. Sentheshanmuganathan, S. and S. Durand. 1969. Isolation and composi- tion of proteins from leaves of plants grown in Ceylon. J. Sci. Food Agr. 20: 603. Shah, F. H. 1963. Thesis, London Univ. Cited by Buchanan (1969) in J. Sci. Food Agr. 20: 359. 1968. Changes in leaf protein, lipids in vitro. J. Sci. Food Agr. 19: 199. Shah, F. H., Riaz-ud-Din and A. Salam. 1967. Effect of heat on digesti- bility of leaf proteins. 1. Toxicity of lipids and their oxi— dation products. Pakis. J. Sci. Ind. Res. 10: 39. Shanley, B. M. G. and O. A. M. Lewis. 1969. The protein nutritional value of wild plants used as diet supplements in Natal (South Africa). Plant Foods Human Nutr. 1: 253. Shany, 3., Y. Birk, B. Gestetner and A. Bondi. 1970a. Preparation, characterization and some prOperties of saponins from lucerne t0ps and roots. J. Sci. Food Agr. 21: 131. Shany, 3., B. Gestetner, Y. Birk and A. Bondi. l970b. Lucerne saponins. 111. Effect of lucerne saponins on larval growth and their detoxification by various sterols. J. Sci. Food Agr. 31: 508. Shenk, J. S. and F. C. Elliott. 1969. .A diet feeder for weanling meadow voles (Microtus pennsylvanicus). Lab. Anim. Care 19: 522. Shenk, J. S., F. C. Elliott and J. W. Thomas. 1970. Meadow vole nutri- tion studies with semi-synthetic diets. J. Nutr. 100: 1437. Shorland, F. B. 1969. Protein fromAwool. Food Manuf. 44 (10): 42. 177 Shurpalekar, K. 3., N. Singh and O. E. Sundaravalli. 1967. Nutritive value of leaf protein concentrates from Medicago sativa, p. 285. In J. Kfihnau (ed.) Physiology and.Biochemistry of'Food Components, Vol. 5 of Proc. Seventh Int. Cong. Nutr., Hamburg, 1966. Pergamon Press, New York. 1969. Nutritive value of leaf protein from lucerne (Medi- Singh, cago sativa): growth responses in rats at different protein levels and to supplementation with lysine and/or methionine. Ind. J. Exp. Biol. 7: 279. N. 1960. Differences in the nature of nitrogen precipitated by various methods from wheat leaf extracts. Biochim. B10phys. Acta 45: 422. 1962. Proteolytic activity of leaf extracts. J. Sci. Food Agr. 13: 325. 1964. Leaf protein extraction from some plants of northern India. J. Food Sci. Technol. 1: 37. 1967. Leaf protein concentrates. Paper presented at Int. Symp. Protein Foods and Concentrates, Mysore, June 24-July 4, 1967. 1969. Leaf protein researCh in India. J. Food Sci. Technol. 6: 165. Singleton, V. L. and F. H. Kratzer. 1969. Toxicity and related physio— Slade, Slade, Smith, Smith, Smith, Smith, logical activity of phenolic substances of plant origin. J. Agr. Food Chem. 17: 497. R. E. 1937. Grass and the national food supply. Rep. Brit. Ass. Adv. Sci. 1937: 457. R. E., D. J. Branscombe and J. C. McGowan. 1945. Protein extraction. Chem. Ind. 1945: 194. A. M. and A. H. Agiza. 1951. The amino acids of several grass- land species, cereals, and bracken. J. Sci. Food Agr. 2: 503. D. 1970. Yield and chemical composition of leaves and stems of alfalfa at intervals up the shoots. J. Agr. Food Chem. 18: 652. D., M. L. Jones, R. F. Johannes and B. R. Baumgardt. 1966. The performance of Vernal and DuPoits alfalfa harvested at first flower or three times by date. Wisc. Agr. Exp. Sta. Res. Rep. 23. E. L. 1941a. The chlorophyll-protein compound of the green leaf. J. Gen. Physiol. 24: 565. 1941b. The action of sodium dodecyl sulfate on the chlorophyll-protein compound of the spinach leaf. J. Gen. Physiol. 24: 583. 178 Smith, E. L. and E. G. Pickels. 1941. The effect of detergents on the ChlorOphyll-protein compound of spinach as studied in the ultra- centrifuge. J. Gen. Physiol. 24: 753. Smith, H. A. 1940. Process for the manufacture of a feed material from lettuce. U.S. Pat. No. 2,190,176. Cited by Kohler et al. (1968) in Tenth Tech. Alfalfa Conf. Proc., Agr. Res. Serv.—74-76, USDA. p. 71. Smith, R. 1969. Leaf protein-~its production and future potential. Food Canada 29 (l): 48. Smith, R. H. 1966. Lipid protein isolates, p. 140. In R. F. Gould (ed.) Wbrld Protein Resources, Adv. Chem. Ser. 57. Amer. Chem. Soc., Washington, D.C. Sollman, T. 1957. A Manual oj’PharmacoZogy and its Applications to Therapeutics and Toxicology, 8th ed. W. B. Saunders Co., Phila- delphia. Spackman, D. H., W. H. Stein and S. Moore. 1958. Automatic recording apparatus for use in the chromatography of amino acids. Anal. Chem. 30: 1190. Spencer, R. R., E. M. Bickoff, G. O. Kohler, S. C. Witt, B. E. Knuckles and A. Mottola. 1969. Alfalfa products by wet fractionation. For presentation at the Amer. Soc. Agr. Eng., June 25, 1969, Lafayette, Ind. Also Trans. Soc. Agr. Eng. 13: 198 (1970). Spencer, R. R., A. C. Mottola, E. M. Bickoff, J. P. Clark and G. O. Kohler. 1971. The PRO-XAN process: the design and evaluation of a pilot plant system for the coagulation and separation of the leaf protein from alfalfa juice. J. Agr. Food Chem..19: 504. Spies,.J. R. 1967. Determination of tryptophan in proteins. Anal. Chem. 39: 1412. Spies, J. R. and D. C. Chambers. 1948. Chemical determination of tryptOphan. Anal. Chem. 20: 30. Stahmann, M. A. 1968a. The potential for protein production from green plants. Econ. Bot. 23: 73. 1968b. The potential for alfalfa protein concentrates in animal and human nutrition, p. 63. In Tenth Tech. Alfalfa Conf. Proc.,Agr. Res. Serv.-74-76, USDA. Strong, F. M. 1966. Naturally occurring toxic factors in plants and animals used as food. Can. Med. Ass. J. 94: 568. Subba Rau, B. H., S. Mahadeviah and N. Singh. 1969. Nutritional studies on whole-extract coagulated leaf protein and fractionated chloro- plastic and cytoplasmic proteins from lucerne (Akmficago sativd. J. Sci. Food Agr. 20: 355. 179 Subrahmanyan, V. and B. K. Sur. 1949. Supplementary value of lucerne to the poor south Indian rice diet. Ind. J. Med. Res. 37: 319. Sur, B. K. and V. Subrahmanyan. 1954. Supplementary value of lucerne at different levels in the poor rice diet. Curr. Sci. 23: 188. Sur, R. 1967. Nutritional studies on leaf protein. Ph.D. Thesis. Univ. Calcutta, Calcutta, India. Cited by Oelshlegel et al. (1969) in J. Agr. Food Chem. 17: 791. Surinder Kaur and P. K. Vijayaraghavan. 1961. Leaf proteins in nutrition. Curr. Sci. 30: 298. Tallarico, G. 1952. (The total utilization of grass.) Ital. Agr. 89: 569. Cited by Raymond and Tilley (1956) in Colon. Plant Anim. Prod. 6: 3. Tangl, H. 1949. (Production of a new type of powdered alfalfa with a high biological value.) Agrartud. 1: 456. Cited by Raymond and Tilley (1956) in Colon. Plant Anim. Prod. 6: 3. Tilley, J. M. A., M. L. Barnes and W. F. Raymond. 1954. Leaf protein extraction. Farm Mech. 6: 481. Tilley, J. M. A. and W. F. Raymond. 1957. The extraction and utiliza- tion of leaf protein. Herbage Abstr. 27: 235. Tracey, M. V. 1948. Leaf protease of tObacco and other plants. Biochem. J. 42: 281. Valli Devi, A., N. A. N. Rao and P. K. Vijayaraghaven. 1965. Isolation and composition of leaf protein from certain species of Indian flora. J. Sci. Food Agr. 16: 116. VanBuren, J. P. and W. B. Robinson. 1969. Formation of complexes between protein and tannic acid. J. Agr. Food Chem. 17: 772. VanEtten, C. H. 1969. Goitrogens, p. 103. In I. E. Liener (ed.) Toxic Constituents of'PZant Foodstufis. Academic Press, New York. VanEtten, C. H., M. E. Daxenbichler and I. A. Wolff. 1969. Natural V glucosinolates (thioglucosides) in food and feeds. J. Agr. Food Chem. 17: 483. VanRiper, G. E. and D. Smith. 1959. Changes in the chemical composition of the herbage of alfalfa, medium red clover, ladino clover, and bromegrass with advance in maturity. Wise. Agr..Exp. Sta. Res. Rep. 4. Waite, R., A. Fensom and S. Lovett. 1953. The amino acid composition of extracted grass-protein. I. The basic amino acids. J. Sci. Food Agr. 4: 28. Walter, E. D., G. R. VanAtta, C. R. Thompson and W. D. Maclay. 1954. Alfalfa saponin. J. Amer. Chem. Soc. 76: 2271. 180 Waterlow, J. C. 1961. The rate of recovery of malnourished infants in relation to the protein and calorie levels of the diet. J. Trop. Pediat. 7: 16. 1962. The absorption and retention of nitrogen from leaf protein by infants recovering from malnutrition. Brit. J. Nutr. 16: 531. Waterlow, J. C. and V. G. Wills. 1960. Balance studies in malnourished Jamaican infants. 1. Absorption and retention of nitrogen and phosphorus. Brit. J. Nutr. 14: 183. West, T. F. 1968. Food for thought. Chem. Ind. 1968: 851. Whitmoyer, T. F. 1956. A laboratory study of growth rate in young Microtus pennsyZvanicus. M.S. Thesis. Mich. State Univ.,E. Lansing, Mich. Wilcke, H. L. 1968. Potential of animal, fish, and certain plant pro- tein sources. J. Dairy Sci. 52: 409. Wilson, A. T. M. 1964. Proc. Sixth Int. Cong. Nutr. (Edinburgh, 1963). Livingstone, p. 336. Cited by Pirie (1966) at the IBP working group meeting, Warsaw, Aug., 1966. Wilson, R. F. and J. M. A. Tilley. 1965. Amino acid composition of lucerne and of lucerne and grass protein preparations. J. Sci. Food Agr. 16: 173. Woodham, A. A. 1965. The nutritive value of leaf protein concentrates. Proc. Nutr. Soc. 24: xxiv. Yang, M. G. and O. Mickelsen. 1969. Cycads, p. 159. In I. E. Liener (ed.) Toxic Constituents of'PZant Foodstuffs. Academic Press, New York. Yemm, E. W. and B. F. Folkes. 1953. The amino acids of cytoplasmic and chloroplastic proteins of barley. Biochem. J. 55: 700. Younus Ahmed, S. and N. Singh. 1969. A heat coagulation unit for leaf protein. Ind. J. TeChnol. 7: 411. APPENDIX Table A1. The principal chemicals used in this study and their sources * Chemical Source In the extraction procedure: Sodium hydroxide Mallinckrodt Chemical Works, St. Louis, Mo. Acetic acid " " " Isopropanol Allied Chemical, New York, N.Y. Acetone Mallinckrodt Chemical Works Hexane Fisher Scientific Co., Fair Lawn, N.J. In the nitrogen and nonprotein nitrogen determinations: Sulfuric acid Fisher Scientific Co. Cupric Sulfate Mallinckrodt Chemical Works Selenium dioxide K & K Laboratories, Plainview, N.Y. Hydrogen peroxide, 30% Mallinckrodt Chemical Works Sodium hydroxide, technical grade purified scales for N determinations " " " Boric acid Fisher Scientific Co. Bromocresol green Matheson, Coleman, & Bell, Norwood, Ohio Methyl red (WS) Nutritional Biodhemicals Corp., Cleveland, Ohio Hydrochloric acid J. T. Baker Chemical Col, Phillipsburg, N.J. p-sulfa-o-methoxybenzeneazo- dimethyl-l-naphthyl- amine Sigma Chemical Co., St. Louis, Mo. Tris (hydroxymethyl) amino- methane, primary standard Trichloroacetic acid Mallinckrodt Chemical Works In the amino acid analyses: Sodium citrate Bio-Rad Laboratories, Richmond, Calif. Thiodiglycol " " Detergent BRIG-35 " " (especially prepared Ninhydrin " " for use in amino acid Cello-solve " " analyzers) Pentachlorophenol Eastman Organic Chemicals, Rochester, N.Y. Standard amino acid calibration mixture Bio-Rad Laboratories _ Norleucine Nutritional Biochemicals Corp. L-cysteic acid monohydrate L—methionine DL sulfoxide Sigma Chemical Corp. L-methionine sulfone " " " Antifoam AF emulsion Dow Corning Corp., Midland, Mich. Potassium hydroxide Mallinckrodt Chemical Works Formic acid, 88% " " " 181 182 Table A1 (cont'd.) * Chemical Source In the tryptophan determination: DL—tryptophan Nutritional Biochemicals Corp. Pronase, B grade Calbiochem, Los Angeles, Calif. p—dimethylaminobenzalde- hyde Eastman Organic Chemicals Sodium phosphate, mono— basic Fisher Scientific Co. Sodium phosphate, dibasic heptohydrate Mallinckrodt Chemical Works Sodium nitrite J. T. Baker Chemical Co. In carbohydrate determination: Phenol J. T. Baker Chemical Co. D(+)mannose Fisher Scientific Co. D galactose Pfanstiehl Laboratories, Waukegan, 111. In lipid determination: Chloroform Mallinckrodt Chemical Works Methanol J. T. Baker Chemical Co. Potassium chloride Mallinckrodt Chemical Works Anhydrous sodium sulfate J. T. Baker Chemical Co. In biological assays: Vitamin diet fortification mixture Nutritional Biochemicals Corp. Salt mix W " " " Alpha-cell, hydrolyzed " " " Casein, hammerstein " " " Dextrin, white technical " " H Sucrose Big Chief, Monitor Sugar Co., Bay City, Mich. Corn starch A. E. Staley Mfg. Co., Decatur, Ill. com 011 H II H H H In organoleptic evaluation: All ingredients were brand names and purchased from a local grocery store. ‘ In high voltage paper electrophoresis: Raw skim milk Mixed milk of Holstein cows that were part of the Michigan State University dairy herd 183 Table A1 (cont'd.) * Chemical Source Pyridine Mallinckrodt Chemical Works Varsol Standard Oil, Okemos, Mich. Ninhydrin General Biochemicals, Chagrin Falls, Ohio 2,4,6 collidine Eastman Organic Chemicals * if a chemical was used more than once, it will be listed only under the first analysis for which it was needed. 184 Table A2. Amino acid composition of alfalfa leaf protein fractions expressed as moles of amino acid/1000 moles amino acids Amino acid Protein fraction 1P3* 2P3 3P3 183 283 383 Lysine 57.17 60.23 79.56 68.13 63.35 56.40 Threonine 56.75 56.94 56.66 63.72 65.03- 60.73 Valine 77.76 81.13 71.81 68.58 59.71 61.00 Methionine 26.30 21.14 20.20 14.74 15.89 8.71 Cystine 12.40 12.90 11.99 34.69 31.89 38.94 Isoleucine 50.67 56.77 48.18 46.88 43.01 43.01 Leucine 86.20 90.19 81.68 61.89 53.60 54.26 Tyrosine 29.63 30.57 34.15 18.42 23.19 20.70 Phenylalanine 48.18 50.99 45.09 29.80 30.64 30.22 Tryptophan 14.78 17.54 14.93 10.51 17.04 15.64 Histidine 26.80 21.91 24.34 16.94 19.07 16.94 Arginine 56.40 53.53 50.71 26.29 28.88 25.51 Aspartic acid 95.69 101.90 94.83 119.04 128.87 138.62 Serine 32.95 39.54 40.54 67.13 74.46 66.47 Glutamic acid 97.43 101.18 91.17 98.25 104.43 113.76 Proline 50.91 52.40 46.82 59.97 66.59 65.63 Glycine 91.80 95.72 86.24 106.04 99.62 104.95 Alanine 88.18 55.41 101.09 88.98 74.72 78.52 * C see methods section, p. 79, for eXplanation of nomenclature 185 Table A3. Amino acid composition of alfalfa leaf protein fractions expressed as grams of amino acid/100 grams sample Amino acid Protein fraction 1P3* 2P3 3P3 183 133 353 Lysine 5.38 4.70 6.09 2.76 1.69 1.39 Threonine 4.21 3.51 3.42 2.04 1.37 1.18 Valine 5.66 4.90 4.25 2.15 1.23 1.16 Methionine 2.53 1.69 1.58 0.61 0.43 0.22 Cystine 0.94 0.81 0.74 1.13 0.68 0.77 Isoleucine 4.21 3.91 3.26 1.68 1.01 0.94 Leucine 7.16 6.22 5.52 2.22 1.26 1.18 Tyrosine 3.55 3.04 3.33 0.95 0.79 0.65 Phenylalanine 5.20 4.57 3.96 1.39 0.94 0.86 Tryptophan 2.02 1.99 1.66 0.62 0.66 0.56 Histidine 2.70 1.83 1.99 0.74 0.54 0.45 Arginine 6.46 5.09 4.73 1.30 0.94 0.77 Aspartic acid 8.08 7.15 6.52 4.34 3.08 3.07 Serine 2.11 2.10 2.11 1.85 1.35 1.11 Glutamic acid 9.23 7.96 7.03 4.02 2.80 2.82 Proline 3.63 3.10 2.72 1.84 1.34 1.23 Glycine 3.84 3.33 2.94 1.92 1.18 1.15 Alanine 4.60 2.40 4.29 2.00 1.10 1.07 * see methods section, p. 79, for explanation of nomenclature HICHIGQN STQTE UNIV. LIBRQRIES III I |lllllfll H 31293102727 72