, 1.1:: 93:; , . . I» 3 ‘ a. . . . . .;v._ r 553.... 4 .. {:3'. n .u . . , .3. .4 Z. .1 . 1.3....1v321 2.. ,.}..Y a... ..,.. .— . 1.. c. 4.. ., . ‘... T: ..::.. 1.5.31.1: . 1.7.1... 25)). .31 v4.'.vb....!4.... v 3 , . v4. .52 x. .14: s 4.; :3... . 4 .~_ 4.11. . .1 ft»; . 2.145133. ...;....».I—.J 7.91.10 1 . I \ . . . .n. .y r... 13.2.... .34. 4...; ,1». ,...:). 1, .4137. 23.7.. a . ..;....y.t. 1.5 v . 945.....4 I: 4:411:13) ...«l. '5'} .1. (£7.14. (v. [.1 , 4 t x. .25.}? . 5.1:...713v u .. f. ......v 2. f:;:.. . .175 F. I. 93.. .. 45.571.71.13») r t...rrtu;v., .2?- ..1.,;)x v I; 1...... A 2. .l...:.:; .5. 1.3%: J. .I .. . rf .v Ziriftlrl. billir). :5 :1 yr): .\ , . .1 ..?»;; ‘ )1.;..>).; .z’vztivnbfritt.) I If; 4):; u. ‘45}, . ; 7.1. 1.5.. 1)... £35.? 4.1....» 1 :I!l:.1lr#zrs:.r:€ :21 :«vwi .35 V . ‘ .4 V. .. :1 .m Inmfiwfiu‘flETJJ 4.4.1.0.; 4.3 1;... .5: s.¥.§x.w:.u=.,fl»..v. HA. if 4. ,. 4 . . A .. . 3.: .4. . .45.; 45.. 6%? «23.. i2 . unknwf‘u.‘ . 4 . . taxi, i... , 73241.3(!) Ann... lllllzllglllMINNylll A I V. ‘4 " v, 3,95“: x . ‘fj-- Weenifytha‘tthe. a. , :g' ' thesis entitled CERTAIN CHEMICAL 'AND NUTRITIONAL ASPECTS_ OF SOYBEAN TEMPE ' ' presented by Slamet Sudarmadji has been accepted towards fulfillment of the requirements for PhD Food Science & degree in Human Nutrition Major professol DMe July 8, 1975 0-7639 Mia?) " V \L."*,‘“\.’~“§ «an, ‘ I 833. H 5.00 0023 f\:\% i/‘\ \ .4 \: x¢~ r. I \ ABSTRACT CERTAIN CHEMICAL AND NUTRITIONAL ASPECTS OF SOYBEAN TEMPE BY Slamet Sudarmadji Tempe was prepared from soybeans and a pure culture of Rhizopus oligosporus. Scanning and transmittance electron micrographs of tempe showed some disruptions of the soybean cell walls and penetration of the mycelium into two layers of cotyledon cells. Based on the liberation of free fatty acids, micro- bial growth and organoleptic changes, the tempe fermentation may be differentiated into three phases: (I) Rapid phase (0-30 hr.), (II) Transition phase (30—60 hr.), and (III) Deterioration phase (beyond 60 hr. of fermentation). It is believed that primarily the contaminating sporeforming bacteria were responsible for the deterioration of tempe. Frying of tempe in coconut oil resulted in a migration of some free fatty acids, but not glycerides, from tempe into the frying oil. Based on the TBA test and organoleptic evaluation, fried tempe, stored for 1.5 months, at room temperature, in bags containing air, did not deteriorate. Slamet Sudarmadji The phytic acid content of soybeans decreased from 1.4% to 1.0% as a result of the tempe fermentation. A strong phytase activity was detected and measured in the tempe and the tempe mold. The pH optimum of this phytase was 5.6 and the Michaelis constant was 0.28 x lO_3M. There were no statistically significant differences in the Protein Efficiency Ratio values (AOAC procedure) between fried soybeans, fried tempe and fried, sesame- supplemented tempe at the 10% protein level in the diet. CERTAIN CHEMICAL AND NUTRITIONAL ASPECTS OF SOYBEAN TEMPE BY Slamet Sudarmadji A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1975 ACKNOWLEDGMENTS The author very sincerely wishes to express his deepest appreciation to Dr. Pericles Markakis for the excellent guidance, encouragement and invaluable assistance throughout the doctoral program toward the completion of this dissertation. The same appreciation is extended to Dr. Georg A. Borgstrom, Dr. E. S. Beneke, Dr. T. I. Hedrick, and Dr. F. R. Peabody for their assistance and suggestions. Special thanks are due to Dr. K. Stevenson and Mrs. Marguerite Dynnik for their help in some microbial aspects of this study, and to Mrs. June P. Mack and Mrs. A. Ackerson for their assistance in electron microsc0py. Last but not least the author wishes to express his gratitude to the Government of Indonesia, the Midwest Uni- versities Consortium for International Activities (MUCIA Inc.) and Gajah Mada University for providing the leave of absence and the financial support for this program. The Department of Food Science and Human Nutrition at Michigan State University has provided ample facilities and excellent cooperation for the whole program. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . LIST OF FIGURES. INTRODUCTION. . . . . . . . . . . . REVIEW OF LITERATURE . . . . . . . . . Village-type preparation of tempe . . . . Mass production of tempe . . . . . . . Tempe mold. . . . . . . . . Mold growth inhibition. . . . . . . . Antibacterial compound. . . . . Nutritional aspects of tempe fermentation Changes in carbohydrates and soluble solids. Changes in protein . . . . . . . . Changes in lipids . . . Toxic materials produced in tempe "bongkrek" MATERIALS AND METHODS. . . . . . . . . Tempe preparation . . . . Sample preparation for electron microsCOpy Extraction of lipids . . . Separation of free fatty acids from glycerides. Diazomethane preparation and esterification of fatty acids. . . . . . . . . Transesterification of glycerides . . . . Deep frying of tempe . . . . . . . Total plate count . . . . TBA method for determination of oxidative stability . . . . . Sensory interpretation of TBA test value. . Phytic acid determination. . . . . . PhOSphorus determination . . . . . . Protein evaluation methods . . . . Procedure for the protein efficiency ratio test iii Page viii 11 12 12 14 17 17 18 20 21 23 27 3O 30 31 32 33 33 36 37 37 38 41 42 44 44 46 RESULTS AND DISCUSSIONS. Electron Micrographs of soybeans and tempe. Chromatography of standard fatty acids Changes in fatty acid profile of tempe during Changes in the fatty acid composition of tempe Rancidity test of tempe using the TBA method Phytic acid content in soybeans and tempe fermentation after frying . . Investigation of phytase activity in tempe. The effect of pH on tempe phytase activity. Determination of Vmax and Km of tempe phytase. Phytase activity in the tempe mold, Rhizopus The protein efficiency ratio values of soybeans, tempe and sesame— supplemented tempe . The supplementation of sesame seed in soybean oligosporus . tempe . . . . . SUMMARY AND CONCLUSIONS. BIBLIOGRAPHY . . . APPENDICES Appendix A. Fatty Acids . . B. Temperature of Tempe at 25°C and 32°C. C. Sensory Rancidity Test D. Animal Feeding Log. E. Standard Curve for Fe. During Fermentation iv Page 49 49 50 63 76 83 88 94 96 99 108 113 123 128 130 138 140 142 143 144 Table 10a. 10b. 11a. 11b. LIST OF TABLES Calorie, protein and fat consumption per capita per day in Indonesia . . . . . . Per capita daily consumption of calorie, protein and fat in selected countries. . . Soybean composition, % on moisture—free basis. Soybeans-production; in thousand metric tons . Strains of RhiZOpus Species which made acceptable tempe. . . . . . . . . . Amino acid composition of tempe and unfer— mented soybeans (by HITACHI amino analyzer). Chemical characteristics of oils from tempe and unfermented soybeans . . . . . . . Distribution of free fatty acids during tempe fermentation . . . . . . . . Water and crude oil contents of soybeans and tempe . . . . . . Free fatty acids in soybeans and tempe, Method I . . . . . . . . . . . . Free fatty acids in boysbeans and tempe, Method II . . . . . . . . . . . Percent distribution of fatty acids among the free fatty acids (ffa) and the glycerides (gly) of tempe (from SP 1000 column) . . Percent distribution of fatty acids among the free fatty acids (ffa) and the glycerides (gly) of tempe (from DEGS—PS column) . . . Page 14 24 25 26 63 64 64 72 73 Table Page 12. Fatty acid composition of soybean oil . . . 75 13. Free fatty acids content of tempe before and after frying in coconut oil at 180°-210°C . 77 14. Free fatty acids content of tempe before and after heating in an oven at 200°C for 20 minutes . . . . . . . . . . . . 77 15. Fatty acid composition of free fatty acid and glyceride portions of tempe, before and after heating in an oven at 200°C for 20 minutes . . . . . . . . . . . . 78 16. Free fatty acids found in coconut frying oil. 79 17. Composition of fatty acid present in the glycerides in fresh coconut oil . . . . 79 18. Fatty acid composition of the glyceride portion of coconut oil, before and after frying of tempe . . . . . . . . . . . . . 81 19. Fatty acid composition of coconut oil . . . 8T 20. Fatty acid composition of the glyceride portion of tempe, before and after frying in coconut oil . . . . . . . . . . . 82 21. Absorbance reading of rancid oil sample in different wavelengths. . . . . . . . 84 22. Sensory test and TBA value. . . . . . . 85 23. Soybeans and tempe samples. . . . . . . 86 24. The TBA value of samples described in the Table 23 . . . . . . . . . . . . 86 25. Phytic acid content of soybeans and tempe. . 89 26. Inorganic phOSphate content of boiled soy- beans and tempe. . . . . . . . . . 93 27. Enzyme activity assay in buffered substrate solution . . . . . . . . . . . . 96 28. Total P liberated in ug and enzyme activity as ug P produced/ml enzyme/hour . . . . 97 vi Table 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Effect of pH on tempe phytase activity in acetate and maleate buffers . . . . . Total production of P in ug per 5 ml reaction mixture. . . . . . . . . . Tempe phytase activity at different pH levels and its percent of maximum activity . . Buffered enzyme—substract solutions . . . . Effect of substrate (phytate) concentration on the P release by tempe phytase in ug/5 ml reaction mixture. . . . . . . Effect of substrate concentration on the initial velocity of phytate- phytase reaction . . . . . . . Lineweaver-Burk plot of the phytate—phytase reaction. Values of l/S and l/V . . . Phytase activities of Rhizopus oligosporus NRRL 2710 grown on PDA, PDA—phytate, PDA— soybeans, PDA-peptone—phytate, PDA—peptone and PDA-coconut liquid and of the mycelia- free medium . . . . . . . . . . . The PER values of casein, soybeans, tempe and sesame-supplemented tempe . . . . . The amino acid composition of an FAO pattern and of eggs, human milk, cow's milk, tempe and sesame. The amino acid requirements of man are also listed. . . . . . . vii Page 98 100 101 103 104 104 107 111 119 125 fl Figure 10. 11. 12. 13. 14. 15. 16. LIST OF FIGURES Products derived from soybeans RhiZOpus mold. Changes occurring during the tempe fer— mentation . Toxoflavin. . Bongkrekic acid Free fatty acid extraction Diazomethane preparation Scanning electron microqraph of boiled soybeans. . Transmittance electron micrograph of soaked soybeans. Transmittance electron micrograph of boiled soybeans. . Transmittance electron micrographos of tempe Scanning electron micrograph of dried tempe. Scanning electron micrograph of soybean tempe The chromatographic profile of standard fatty acid mixture on a 10% The chromatographic profile of standard fatty acid mixture on a 10% SP 1000 column DEGS—PS column Total free fatty acids produced during fermentation viii Page 16 22 28 29 34 35 51 52 53 54 56 57 58 62 66 Figure 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Total bacterial count in tempe during fermentation . . . . Transmittance electron micrograph of con- taminating bacteria in tempe. The temperature of tempe incubated at room temperature and in an incubator at 32°C Fatty acid composition (in %) of free fatty acid and glyceride portions of tempe during fermentation. . . . . Inositol hexaphosphate (phytic acid). Two possible structures. . . . . . . . pH Optimum curve for tempe phytase . . . Effect of phytate concentrations on tempe phytase activity. . . . . . . Lineweaver—Burk plot for the tempe phyate— phytase reaction. . . . . . . Average weight of rats in each of four groups fed casein, tempe, soybeans and sesame— supplemented tempe basal diets at 10% total protein in the diet during 28 days. Weight gain (g) of rat groups (6 rats/group), fed different diets, at 10% total protein in the diet . . . . . . . . . . 1x Page 67 68 7O 74 90 102 105 106 117 118 INTRODUCT I ON In recent decades the explosive growth of world population, the accelerating depletion of natural resources, and the severe food scarcity in large areas of the world have placed new emphasis on the production, storage, transportation, processing, preservation and distribution of foods, as well as on the study of their nutritional qualities. According to the Food and Nutrition Board, National Academy of Sciences, National Research Council, the recommended daily dietary allowance for adult American male is 2800 Cal. and 65 g of protein. For small bodied South East Asian people, working manual labor in 25°C climate, FAO has computed at between 1625—1821 Cal. per person per day, and minimum crude protein requirement is 24.5 g (Latham et 31., 1972; Clark, 1972). Meanwhile for a slightly bigger East Asian adult man (body weight 55 kg), joint FAQ/WHO expert groups computed at 2530 Ca1./day (FAQ/US HEW, 1972). For women the caloric requirements are about 25% less. The Food Consumption Table compiled by FAO (1971) from 132 countries, has revealed that Ireland and New Zealand are two countries with the highest average calorie and protein consumption in the world, 3455 and 3454 Cal. and 91.5 and 109.6 g protein per capita per day, respec— tively. On the contrary, the Indonesian people consumed the least food in the world with only 1760 Cal. and 38.4 g protein per person per day, which stand at the precarious border of hunger even by the South East Asian standards. Furthermore, Table 1 shows that most of the foods consumed by the Indonesians originate from plant materials. Table l.--Calorie, protein and fat consumption per capita per day in Indonesia (FAO, 1971). Calorie Protein Fat (total) (9) (9) Vegetable 1714 (97.4%) 33.2 (86.5%) 20.4 (89.5%) Animal 46 ( 2.6%) 5.2 (13.5%) 2.4 (10.5%) Total 1760 (100%) 38.4 (100%) 22.8 (100%) For comparison, Table 2 shows the calorie, protein and fat consumption in some selected countries. The protein and calorie malnutrition may be critical to child development. Besides the well known kwashiorkor and marasmus syndrome, protein deficiency more importantly will affect the brain development especially in fetal and early life of a child. Besides zinc and vitamin B6, Table 2.-—Per capita daily consumption of calorie, protein and fat in selected countries* (FAO, 1971). Protein (g) Fat (g) Country Cal/day per day per day Afghanistan 2057 65.4 26.0 Albania 2366 71.3 39.1 Argentina 2885 90.2 105.6 Australia 3130 90.9 129.2 Bolivia 1765 45.8 33.9 Brazil 2541 63.9 51.3 Burma 2011 44.1 30.5 Canada 3142 94.5 141.4 China 2045 58.2 31.2 Columbia 2192 50.1 47.3 Congo 2160 39.8 37.5 Cuba 2501 62.8 43.7 France 3108 98.2 136.6 Germany (East) 3040 76.4 129.0 Germany (West) 2927 80.1 133.8 Greece 2901 97.7 94.8 Honduras 1930 48.6 37.6 Iceland 2899 98.8 131.6 India 1964 50.6 28.2 Indonesia 1760 3814 22.8 Iran 2029 55.2 39.0 Ireland 3455 91.5 135.8 Japan 2416 72.4 41.3 Malaysia 2200 49.4 40.8 Mexico 2624 66.5 58.2 Mongolia 2538 92.9 81.7 New Zealand 3454 109.6 154.0 Pakistan 1995 49.6 28.6 Philippines 1911 43.8 31.4 Saudi Arabia 2082 56.2 29.9 Singapore 2443 63.3 48.9 Taiwan 2379 58.7 45.2 Thailand 2226 50.9 27.3 UK 3233 88.6 141.9 USA 3156 93.7 148.0 USSR 3182 92.2 74.5 *Figures do not include beverages. proteins act as the backbone of lipid incorporation in the brain, and therefore a protein deficit will cause a longer lasting deleterious effect on the brain than a transient deficit in essential fatty acids and calorie. The critical period for brain development in man begins with the second half of fetal life and ends about 18 months after birth. Apathy of the children is one of the most constant signs of protein—calorie malnutrition; the loss of curiosity and the lack of desire for exploration, a progressive withdrawal from the environment are other symptoms of acute protein deficiency (Cheek, 1968; Muralt, 1972; O'Neal et 21., 1970; Anonymous, 1972). The provision of adequate proteins and nutritious foods for the people is of prime importance for the present and future generations. Besides the promising source of food from the ocean, it is obvious that foods from plant origins play a very important role in the diet of Indonesians as well as the peoples from other developing countries, because of their immediate availability and traditional acceptance. One of the most promising plant foods is soy- beans due to its acceptability (eSpecially in the Orient) and the source of good quality proteins. The protein, fat, carbohydrate and ash contents of soybeans are presented in Table 3. Protein analysis of soybeans (Table 6, p. 24) shows the composition of amino acids which contains all the essential amino acids, i.e., Lysine, leucine, isoleucine, Table 3.--Soybean composition, % on moisture-free basis (Cowan, 1969). Assayed Protein (Nx6.25) Fat Carbohydrate Ash Whole bean 40 21 34 4.9 Cotyledon 43 23 29 5.0 Hull 8.8 1.0 86 4.3 methionine, phenylalanine, threonine, tryptophan and valine. Methionine is present in relatively low concentration. A plant protein may be utilized by humans best when it is consumed directly without converting it into animal protein. Through conversion, soybeans yield about 43 lbs of edible animal protein per acre, while through direct consumption they yield about 600 lbs of protein per acre (Spaeth, 1974). Through the advancement of soybean technology, it is feasible now to convert soybeans into many kinds of products which have higher acceptability as food than the unprocessed soybeans. Many soybean base products are already available in the market and others are being developed. The Figure 1 shows the varieties of soybean base products. The Association of South East Asian Nations (ASEAN) on their meeting in Jakarta has recognized the importance of soybeans as food and recommended "further studies to improve soybean processing techniques to overcome protein ou-waom vac: mcodmaaaa uaooc meanssanuum m~40qauzu noaquxoa mdauma vcuonrom muuuuamwn xaqa uqdu Aucqcm. nucoaoqm meauoeuou ucuod mauuuoa Magnum novauauuuuCu 1cm aca-a 4com< ocdnuomuao ~0:ou«< anew» scout IdOm-wu:< onuwuu cum meacuuuonm ucoo< ocqnaaqnqum ocauaOuax ucoo< m:4uouuammuwu:¢ on: xuduUan mm: aaoauuz -c0auuuusz manuauaouususcn nocfluQOU aunaouonu muovlom omonu>on uCoo< o>quu¢ oouuudm nuunvOum aucnu nuusv0um auoxdn acoo< ocdxcuoaaaa Unconaaaz ucmaoo uOOuQnouuz QaOm >uuam nuwuauaumuan moc4uaoo o>auu~u0ua axCu v:«u:aun nonundm nudge ocqxuom suwaocqq meHUSDUUmcH coauoasmc~ ~au«uuuu~u nucuuooucanuo naao ouoo avCSOQEOU ocdxasuu maucauduwr mcanoaawx ocnc0uho£m wanouooo> muvmumm :0.)v:1m .d‘o cad-m cocdmnouo unaum adsvduaounehacn ocduaonur uuuqccoxaz nuao meaxooo .uuz u.uognauc4 mom: o~n4vu _ ac‘anoh caducdoc sW—uvahn hUAuh H00: odnnuooo> uvnuh henna onuuxop ocauam mucasm co..~:lu noun: ocduaou human m>.mo£u£ can: ~aquun20c~ _ mucus xuuuoio mCOAufl—JEHON UCGMCM n4:4hv voou xfluz sow muoauu>on mucmuc0u u noun-um Joaaufizuudt noun cottage anew: ocauouo< muqn ”nos vououvxzuo memos coauau‘aa new mconqu canm auuavOua coo: veg-Hanan 03~o> o>dughuac o>ounfia ecu acoucou caoDOuQ onQOHQCM ck o>iuavv< voou mono angina _ caouOua >00 vaunfio-H naoun xnauoao omen ovamm nauusn u3c>om oeuuou >om scodvrnocu noxu-uu non: quantum? noun—odnavu non: ~ooucdook _ _ :qcuqooa cnon>om aao clonxoo sauna coco-no: aouvznoooe Haney-qum «ouuunqloaum Houoo>~u coda. >uu-h naououm F 1 _ noauu>auoo :don>0m dao cevn>om ovshu naustPm Ado undamoa Idxooo acodeuocu oaxoou madam ~¢ucowuo rIIIIIL nc10a>0m voua60¢ coauuoucoo ucoavoumcu >ocuu anon“ _ . 2 _ mausnOHA coonxom nchuo ngoauo unou-)Qu x8 fiesta rlixvlucllltxl.:ttuarl-. nuax undue" noooo uoolm unauo can hoficvulu Umuouo Cam Macaw oluocos nuuounoo caucum a»: unczosoo x6540 nacho avoom xUOum maneumm cawn>om ovum acovnxom voxam _ _ nuuauoha yam dash Hausuaz m2¢mn»0m nuauo muduoanauc4 mad-dua> noduudu houa~quhom «on: "nauuunvCu p .thma .zuuuvad new uuoum. ncdun>oa lawn oo>quov ouuavoum muoom you moucuuaoucoo :«Ououm swoon P3140.“ avoom xuouno>u4 noun noon h I "col c40A>Om _ o>auqvvn uaoz acoauOuvc4 >uoxfln wuohucoucou cabuOHQ >om hi manna waivednduct caadua> nouuudo ova...— uunau .353: _ 0.0m Add! «an m noon educanquc< umao> uneduuaz mucfiam ousuxok ocqxuun u=o~OCwA aucoauu ucuon vane canon Onuaunum mucumn “undouuuoacu vuaonaaua vooaham u>auuzu¢ WUQD alukdmflflCH P muoavOHQ us: .uqanu L .d august .unua vOuQHSIam cuououa oHn-uooo> vousuxue c»0u0»m smut o‘cuvuu-< u.uonouo mecca uuflo aqwuomm muu3u0um >vcnu mcoauuwucou waa oucwmuoaavuoa>r voccnu mmunuucuucoU >no voou snum muwvaom ooauo>om mama mauonoooom mauunJ u mxmucmm mona: uwhflmwuu wmhuuuo: unmnOustmmm mgmwuwu wcumu :ou «Aflzu mwauuma uuMi vmamanm quMAUUmm muonUOum vmmz mvmuooz moumou hnuucwaud< vqoum ucoquumca huuxun mom: can‘wu _ nuquo vac uncaw >0m _ nuusvoum oxdam flouuuuoo b deficiency problems" (The Strait Times of Singapore, May 10, 1974). Soybeans are widely cultivated in South East Asia and already play an important part of people diet for centuries. The largest soybeans producer among the ASEAN members is Indonesia, which is also the fifth largest soybean producer in the world after the USA, China, Brazil, and the USSR (United Nations, 1974, Table 4). Some of the new soybean products such as textured soy protein, simulated meat etc., require a high degree of technOIOgical know-how and big capital inputs which are lacking in most developing countries at the present time. Therefore traditional soybean fermentation processes which require only simple equipment and can be done at the village level, are the most feasible answer. Sufu, Hamanatto, Shoyu, Miso and Tofu are soybean products of China and Japan, while Tempe and Kecap are products of Indonesia which owe their popularity to the simplicity in processing and better acceptability as foods and condiments than the plain unfermented soybeans. Tempe has been one of the most important traditional foods in Indonesia for centuries. However, the traditional village-type of tempe fermentation process lends itself to some improvements, such as in the area of sanitary con- ditions, the use of a desirable and purified mold culture stocks and finding a way for simple preservation methods for a delayed consumption time. The implementation of those improvement programs will require a tedious education Table 4.--Soybeans production; in thousand metric tons (United Nations, 1974). Country 1966 1967 1968 1969 1970 1971 1972 WORLD 39,084 40,709 43,997 45,180 46,525 48,680 53,289 Argentina 18 21 22 32 27 59 78 Australia 1 l 1 1 6 9 26 Brazil 595 716 654 1,057 1,509 2,218 3,666 Canada 245 220 246 209 283 280 375 China 10,970 11,100 10,670 10,920 11,580 11,680 11,500 Colombia 52 80 101 100 96 83 80 Hungary 1 0.0 0.0 0.0 0.0 0.0 0.0 Indonesia 417 416 420 389 498 475 515 Japan 199 190 168 136 126 122 127 Khmer Rep. 7 8 4 9 4 2 3 Korea, Dem. People's Rep. 215 225 240 225 228 230 235 Korea, Republic of 161 201 245 229 232 222 224 Mexico 95 106 275 266 280 250 366 Nigeria 52 60 54 62 58 63 63 Paraguay 20 18 14 22 3O 74 128 Romania 20 41 47 51 91 165 186 South Africa 3 4 4 7 3 3 4 Thailand 38 53 45 61 50 54 56- Turkey 5 6 9 ll 12 ll 13 USSR 586 543 528 434 604 535 580 United States 25,270 26,575 30,127 30,839 30,675 32,006 34,916 Viet—Nam, Dem. Rep. of 14 17 17 18 19 19 19 Viet—Nam, Rep. of 8 6 7 6 7 8 7 Yugoslavia ll 9 3 5 5 4 6 efforts in health and nutrition for the people, a favorable economic condition to acquire an improved new technique and most be supported by research and experiments in chemistry and preservation technology. One alternative for improvement is a centralized tempe fermentation plant, by setting up modern and scientifically controlled plants in some key locations. At this time, however, the ex.ent of the impacts of this endeavor must be carefully studied to the employment and the livelihood of many small traditional tempe producers. This present study on tempe is only one small part of the never-ending assignments of scientific quest and better understanding of our food for more efficient utili— zation of this limited resource. On his essay review, Hudson (1974) wrote that lipid is the first group of the major naturally occurring bulk organic compounds, which yields the battery of elegant techniques, especially gas chromatography, providing an analysis with a more precise result in a few hours. Lipids are the key to many biological processes, whether they are chloroplasts or erythrocytes, or the structure of brain and nervous sytems. In the preservation of foods, lipids play an important role also. The development of off flavor during storage or processing of foods has been attributed to the degradation of lipids, either catalyzed by enzymes or initiated by air oxydation. 10 It is now recognized that oils and fats play a far more important role than merely as sources of energy. Nutrition aspects besides the aesthetic, pleasing properties and wholesomeness are of prime importance in foods and should serve as the main objective directly or indirectly in food research. In this study, the tempe fermentation is explored, especially on its effect to the lipids, protein and phytic acid. Phytic acid has drawn much attention in the last few years because of its metal binding properties capable of inducing trace element deficiencies in human and animals. The tempe fermentation has been proven to improve the nutritional value of soybeans. Hopefully, this present study will contribute to the understanding of tempe fer- mentation and thence the appreciation of this ancient and remarkable fermentation practice. REV IEW OF LITERATURE Tempe (tempeh) is an Indonesian food generally made from soybeans fermented with Rhizopus oligosporus Saito and other Rhizopus species. It is a cake-like solid mass of soybeans held together by mold mycelia, whitish in color and never eaten raw but usually cooked as soups, deep fried in oil or fried as chips. This fermented soybean food is eaten by millions of people in Indonesia and Surinam, and some tempe is manu— factured in Holland. Hesseltine (1965) reported that tempe is highly acceptable to Americans and Europeans. For cheap substitution of soybeans or simply as another variation of tempe, copra cake (bungkil, bongkrek) can be used solely or mixed with soybeans. The fermented product is known as tempe bongkrek. Other types of tempe such as tempe koro made from Phaseolus beans, tempe gembus /' " made from soybean curd (tofu)(whey)are also popular. Another product similar to tempe called oncom is made from pressed peanut cake (as peanut oil by—product) and fermented with Neurospora sp., has a yellowish pink color rather than whitish as in tempe (Steinkraus g3 g1., 1965a). 11 12 Hesseltine g5 g1. (1967) successfully introduced new fermented tempe-type products prepared from wheat, oat, rye, barley, rice and a combination of rice or wheat with soybeans. The new products possess a very pleasant odor, a desirable color and a very acceptable mild taste. Village—type preparation of tempe Dry soybeans are soaked overnight in tap water and then placed in a bamboo basket at the edge of a stream so as to float the seed coats away while the beans are trod on with the feet. The dehulled beans are then boiled without pressure in excess water for half an hour. The beans are removed from the cooking water thoroughly drained and cooled. After cooling and draining, the beans are inoculated with spores of Rhizopus strains. The inoculated beans are tightly packed and wrapped in banana leaves about 1 cm thick, in various shapes and sizes. The banana leaves are believed to contribute to a desirable flavor. Some forms of tempe are fermented in bamboo cylinders. The Rhizopus inoculum is taken from pieces of a previous fermentation cakes or from the wrapper, and also commercially available as dry spore preparations on dry Hibiscus tiliaceus leaves. Mass production of tempe Martinelli g3 g1, (1964) reported a method to make tempe rapidly in large amounts by pure culture fermentation 13 in shallow wooden and metal trays with perforated bottoms and covers. Excellent tempe was also made in perforated plastic bags and tubes. For plant-type mass production of tempe, Steinkraus gg g1. (1965b) described a pilot-plant process for the production of dehydrated tempe. The hydrated beans were passed through a burr mill and the hulls were removed by floatation with water or through gravity separator. The hydration was accomplished by soaking the dry beans in water acidified with lactic or acetic acid to inhibit bacterial growth overnight at room temperature. The beans were precooked at 100°C in water, and then drained and cooled. As the temperature fell to 35—38°C, they were inoculated with lyophilized tempe mold, three grams per kilogram of precooked beans. The inoculated beans were mixed for five minutes in a bowl with a Hobart mixer. About 3 kg of inoculated beans were spread on a 35 x 81 x 1.3 cm stainless steel tray with 3 mm mesh bottom, covered with waxed paper to decrease moisture loss and were incubated at 35-38°C in 75—85% relative humidity for 15-18 hours. Wang gg g1. (1975) successfully mass-produced Rhizopus oligOSporus spores on rice, combination of rice: wheat bran or wheat: wheat bran, then freeze—dried and ground into fine powder for tempe inoculation. 14 Tempe mold Village-style fermentation of tempe is usually done with impure cultures of molds. Stahel (1946), Van Veen and Schaefer (1950) and Steikraus gg g1. (1960) stated that Rhizopus oryzae Went and Geerligs is the mold chiefly responsible for tempe fermentation. Later Hesseltine (1965) isolated 40 strains of Rhizopus from tempe which could produce an acceptable product when soybeans were inoculated with these strains (Table 5). Table 5.--Strains of Rhizopus species which made acceptable tempe (Hesseltine, 1965). Name No. of Strains RhiZOpus oligosporus Saito 25 R. stolonifer (Ehren) Vuill 4 R. arrhizus Fischer 3 R. oryzae Went and Geerligs 3 R. formosaensis Nakazawa 3 R. achlamydosporus Takeda 2 Total 40 Hesseltine concluded that obviously R. oligosporus is the principal species used in Indonesia for making tempe. Boedijn (1958) described R. oligosporus as a mold which can always be isolated from tempe cakes, bungkil (cattle cakes) and fermenting tobacco. 15 General description of Rhizgpus (Frazier, 1957). Systematical classification: DIVISION: Thallophyta SUBDIVISION: Eumycetes (true fungi) CLASS: Phycomycetes (nonseptate) SUBCLASS: Zygomycetes (sexual spores are zygospores) ORDER: Mucorales GENUS: Rhizopus Genus Rhizopus has distinguishing characteristics (Figure 2): 1. nonseptate 2. has stolons and rhizoids, often darkening in age 3. sporangiOphores arises at the nodes where rhizoids also are formed 4. Sporangia are large and usually black 5. hemispherical columella and cup-shaped apOphysis (base to the sporangium) 6. abundant, cottony mycellium 7. no sporangioles R. oligOSporus can use materials such as xylose, glucose, galactose, trehalose, cellobiose and soluble starch, soy- bean oil, but not raffinose. It can use aSparagine and ammonium sulfate as sources of nitrogen. It also produces a protease and lipase (Wagenknecht gg g£., 1961 and Hesseltine, 1965). Oxygen is essential to the growth of the mold. Steinkraus gg g1. (1960) and Martinelli gg gi. (1964) have 16 Sporangium . sporangiospores Q ~ columella apophysis . node sporangiophores // u u ,flva” rh1201ds Igggggggj stolon fl‘fifi’ Figure 2. RniZOpus mold (Frazier, 1957). demonstrated that when the layer of fermenting beans was thicker than 2 inches (5 cm) mold growth became less rapid and less heavy in the center than it was in the thinner layer. To make a good tempe they recommended that the thickness should not be more than 3 cm. To get the right amount of aeration, perforated metal trays or plastic bags can beused. However, if the amount of aeration is in excess, the soybeans at the surface will dry out before mold starts to develop, and produce undesirable black spores (Martinelli g: gi., 1964). The tempe mold grew well at 31—37°C and requires 22-24 hrs. to full fermentation. At temperatures below 30° and above 44°C soybeans fail to ferment properly. The length of time to full fermentation depends on the temper— ature of incubation. At 25°C it requires 80 hrs. and at 28°C requires 26 hrs. 17 If the pH falls below 3.5 the growth is slow. Mold growth inhibition The important step in making tempe is to boil the dehulled soybeans in excess water and then drain and cool prior to inoculation. If dehulling and draining are not applied the mold does not grow well, a lot of spores are produced and the tempe does not have a pleasant flavor (Hesseltine, 1965 and Steinkraus g3 g£., 1960). Hesseltine, g3 gi. (1963) observed that an in- hibiting substance exists in all 16 varieties of soybeans tested. This inhibitor must be soluble in water and heat stable. Another possibility is a growth promoting substance for bacteria might remain in the soybeans and in turn restricts the growth of fermenting mold. Addition of tetracycline to the beans and lack of bacterial evidence from samples grown on nutrient agar, proves that the latter possibility is unlikely. Antibacterial compound A substance having bacterial inhibiting activity has been isolated by Wang g3 gl. (1969) from tempe. This substance, probably consisting of four or five components, inhibits the growth of some Gram positive and Gram negative bacteria. This antibacterial substance is produced by R. oligosporus and behaves like many antibiotics which at low level stimulate bacterial growth and at higher level inhibit their growth. 18 The antibacterial activity of crude tempe extracts is rapidly destroyed at room temperature. The isolated product, on the other hand, is fairly stable to boiling and in the pH range of 2 to 7. This substance perhaps contri- butes to the resistance of Indonesian people to diseases. Nutritional aspects of tempe fermentation The plain soybeans are almost unpalatable to the Indonesians, and consumed only in small amount as snacks, soybean milk (saridele) and bean curd (tofu); soybean sprouts are also used. Some factors such as the difficulty to digest in soybeans insufficiently cooked (antitryptic factors), their undesirable beany flavor and flatulence which may accompany their ingestion make plain soybeans unpopular as food. The high phytic acid content in raw soybeans may also have adverse nutritional effects, by making certain minerals unavailable to the body. However, when soybeans are made into tempe, they are easy to digest, have a pleasant odor, and even they are suitable for people suffering from dysentry and nutritional edema (Van Veen and Schaefer, 1950). In 1974, the procedure to improve soybean flavor was patented in Japan by immersing beans in a liquid containing a ribonuclease from RhizoEus (Toyo Spinning Co., Japanese Patent, 1974). There must be some desirable changes to the soybeans during tempe fermentation by RhiZOpus mold. 19 GyOrgy (1961) studied the tempe nutritionally as compared to soybean flour and skim milk. After a ten week feeding period of rats, identical weight gain and equal protein efficiency were noted for tempe, soybean flour and skim milk at the 20% protein level. In contrast at the 10% protein level the PER of plain soybeans was smaller than that of skim milk, whereas tempe was as good as skim milk. The superiority in nutritive value of tempe over the control plain soybeans is significant only at low protein intakes. Some rats receiving diet containing 10% protein, all from soybean flour, died of severe massive hemmorrhagic necrosis of the liver (severe vitamine E deficiency), and the rest of the rats receiving the same ration suffered from cirrhosis of the liver or acute necrotizing nephrosis. Rats with tempe as well as with skim milk diet both at 20% and 10% protein have shown at autopsy normal livers and kidneys. The prevention of rancidity in tempe and the pre— vention of vitamine E deficiency symptoms in rats receiving tempe can be traced to the presence of an "antioxidant." It has been reported that soyflour developed strong rancidity early after exposure to air at room temperature, in contrast with tempe under identical conditions remain free from organoleptic rancidity even after 1.5 years. In 1964, Gyérgy gg gl. isolated an active con- centrate extracted from tempe with alcohol or ether showing 20 a definite protection in hemolysis test, but did not correspond to d—toc0pherol or to F84 which was isolated from yeast. Steinkraus gg gl. (1961) also noted that the superiority of tempe PER is damaged through drying at 150°F. Loss of solids and protein in dehulling, soaking, washing and cooking of soybeans did not reduce the nutritive value of tempe as already demonstrated by Smith g5 gi. (1964). They further proved using rats as experi— mental animals that methionine supplementation of tempe significantly increased rate of growth and protein effi— ciency values. About one-third of the thiamin of soybeans was used up by R. oryzae was reported by Roelofsen and Talens (1964) but riboflavin, niacin, and vitamin B6 increased consider— ably to 5—6 times higher than the unfermented soybeans. Vitamin A was reported about the same as in raw soybeans (Jansen and Donath, 1924; Murata, 1965). A nutritional improvement project for children in develOping countries using tempe as a food source has been initiated by FAQ. A very promising large scale tempe feeding experiment has been reported in Southern Rhodesia (Autret and Van Veen, 1955). Changes in carbohydrates and soluble solids Soluble solids, soluble nitrogen, fiber, nitrogen— free extract (NFE) and ammonia are increasing during tempe 21 fermentation. The temperature in the fermenting bean mass is rising above the incubator temperature, reaching maximum after 30 hours and then declines. Steinkraus g3 gi. (1960, 1961) observed that soluble solids rise from about 13% to nearly 28%. Soluble nitrogen rises from about 0.5% to nearly 2%, while total nitrogen remains about 7.5%. The pH also rises during the fermenta— tion (Fig. 3). Van Buren g3 g£., (1972) reported an increase in fiber, nitrogen free extract (NFE) and ammonia. They also suggested that the increased water—soluble NFE, in contrast to the decreasing 66% ethanol—soluble, NFE, is due to water soluble pectic and hemicellulose—type material solubilized by mold enzymes and apparently responsible for the softening effect of the mold on the cooked soybeans. Non enzymatic browning during fermentation and drying was also reported. Steinkraus g3 gi. (1960) also made cytological studies on tempe and concluced that the cells of beans cooked in water and the tempe made from them remain intact after blending in a Waring blender while soybeans cooked in steam for 90 minutes contained relatively fewer intact cells after similar blending. Changes in protein The ammonia content rises during the course of tempe fermentation suggesting deamination of amino acids. 22 II II .Aamma ..Hm pm msmuxcflmumv coflumpc®Eme demp 030 mcHHSU mcfluusooo mmmcmnu .m ousmflm mndom mm om me om ma _ . _ _ _ a I a q . _ _ d o.m 40m % o.w%mm I a m - w m m mOOGMDmbdm o.n%om m mcflospmm rd m z mannaom "D I .u D m a 0m 0 s o v UA an H m 0 o.m:mm mg "- ousumnmdaoa no "pcwmmq tow spIIos qunIos (%) 23 Experiments by Van Buren g: gl. (1972) showed the increasing water soluble protein during the active mold growth. The disappearance of crude protein, soluble in 66% ethanol, shows that the mold utilizes amino acids and low molecular weight peptides, and probably the intermediate- size protein breakdown products contribute to the water soluble protein. After growth ceased there was an accumulation of low molecular weight nitrogen compounds. It was also noted that the solubilization of protein was accomplished without the development of bitter materials as in other cases of enzymatic proteolysis of soy protein. Murata (1965) concluded from his chemical analyses of tempe that the protein, amino acids, moisture, crude fat, ash and fiber contents are not much different from those of unfermented soybeans. The amino acid composition of soybeans and tempe appears in Table 6. Changes in lipids Wagenknecht gg gi. (1961) and Van Buren gg gi. (1972) investigated the changes in soybean lipids during tempe fermentation using R. oryzae mold. One of their conclusions was that the concentration of total ether- extractable lipids rose slightly at the time of most active mold growth (20—30 hrs) and then diminished. The acidity increased, but at the same time the pH also showed a steady increase throughout the fermentation time. This is presumably due to liberation of ammonia or other basic 24 Table 6.-—Amino acid composition of tempe and unfermented soybeans (by HITACHI amino acid analyzer) (Murata, 1965). Unfermented Amino Acids Soybeans (mg/gN) Tempe Aspartic acid 744 756 Threonine 278 282 Serine 270 268 Glutamic acid 1049 1001 Proline 342 309 Glycine 292 275 Alanine 250 228 Cystine (M) 113 121 Valine 328 345 Methionine (M) 77 81 Isoleucine 338 356 Leucine 525 565 Tyrosine 171 161 Phenylalanine 302 302 TryptOphan (M) 67 87 Lysine 392 410 Histidine 160 167 (NH3) 140 169 Arginine 491 440 Nitrogen % 7.85 8.06 Note: (M) = microbiological assay using Leuconostoc mesenteroides. 25 end products of protein. The chemical characteristics of the oils of soybeans and tempe are presented in Table 7. Table 7.--Chemical characteristics of oils from tempe and unfermented soybeans (Murata, 1965). Unfermented Soybeans Tempe Acid value 1.02 50.59 Refractive index nés 1.4730 1.4711 Iodine value 128.5 126.1 Saponification value 191.1 189.2 Composition of fatty acids: Cl6:0 10.4% 10.9% Cl8:0 5.1% 4 9% C18:1 26.8% 28.1% Cl8:2 50 0% 49.4% C18:3 7 8% 6.8% The lipase activity of R. oryzae is also increasing during fermentation time. About one—third of neutral fat of soybeans is hydrolyzed by the fungal lipase, but there is no subsequent utilization of the liberated fatty acids. The fatty acid content of tempe during fermentation is shown in Table 8. Steinkraus g2 g1. (1961) and Gyorgy (1961) detected and isolated an antioxidant produced during the tempe fer— mentation. They determined the peroxide values of dried, pulverized and stored soybeans ranged from 18.3 to 201.9, 26 Table 8.-—Distribution of free fatty acids during tempe fermentation (Wagenknecht g3 g£., 1961). Mg/lOO g tempe C16z0 18:0 C18:1 C18:2 C18:3 TOtal Cooked soybeans neutral fat . . . free fatty acids 41 31 127 0 0 199 24 hour tempe 420 175 713 2510 293 4111 30 hour tempe 771 202 802 2543 204 4522 48 hour tempe 665 202 1359 4138 304 6668 69 hour tempe 863 367 1671 5032 302 8235 Mol percent Cooked soybeans neutral fat 10.86 3.90 18.11 61.00 6.13 free fatty acids 22.24 15.21 62.53 0 0 24 hour tempe 19.81 3.75 15.39 54.62 6.43 30 hour tempe 18.23 4.31 17.24 55.08 4.45 48 hour tempe 10.83 2.96 20.07 61.57 4.56 69 hour tempe 11.37 4.36 19.98 60.62 3.67 Cooked soybeans 24.hour tempe 30 hour tempe 483h£>ur tempe 69 110 ur tempe \ Total free fatty acids g/100 g tempe 0.26 3.59 4.77 6.93 8.19 Percent free fatty acids of total ether extract 1.09 13.87 18.93 30.00 35.11 27 while the peroxide values of tempe under identical con— ditions ranged from O to 1.1. Gyérgy g3 gl. (1964) isolated by paper chromato- graphy a compound (Rf=0.92) which proved to prevent the hemolysis of red blood cells of a vitamin E deficient rat with dialuric acid, as well as giving a positive ferric chloric reaction and Emmerie-Engel test. This compound was subsequently identified as 6,7,4' trihydroxyisoflavone, which has Amax of 260 and 327 nm, and also is a normal constituent of unfermented soybeans. This antioxidant can be liberated from raw soybeans by hydrochloric acid hydrolysis. The hypothesis was that the liberation of trihydroxyisoflavone will protect the biologically effective vitamine E presents in soybeans. The effect of trihydroxyisoflavone itself in vitamine E deficient animal is still unknown. Toxic materials produced in tempe "bongkrek" Soybean tempe or locally called tempe kedele is never reported as the cause of any kind of food poisoning. However, if the soybeans are mixed with coconut cake (copra cake) or prepared entirely from copra (for a cheaper, low quality tempe), the product bongkrek may occasionally become poisonous. These bongkrek poisons are very potent toxins. People die or become seriously 111 after a consumption of the poisonous bongkrek even in small quantities (Van Veen and Mertens, 1934; Van Veen, 1935, 1936). 28 The occasional toxicity of bongkrek is attributed robic bacterium growing to Pseudomonas cocovenenans, an ae as a contaminant during the bongkrek fermentation. oflavin and bong— Two primary substances, namely tox krekic acid, are responsible for the bongkrek toxicity (Figures 4 and 5). O l N H3C”N/ \c// \ 1 l CI 4C C O \\N/’\\\N’// CH3 Figure 4. Toxoflavin (Van Damme gg g;., 1960). 29 .Amnma ..mm mm cmflsum 00v Uflom oexmuxmcom .m musmflm 3000 mm \ mmU U m m/ \ m mm m mmo m m \m mo I o \ / \ /o / .I 4 m / I10 U\U” \oflo \ /m\o/ /oH u\ /mm\ o m _ P / U H \ / U E 0003 mooo m \ o o m m moo m m m mo MATERIALS AND METHODS Tempe preparation The tempe was prepared by a method similar to that practiced in Indonesian villages and described by Hesseltine (1965). The soybeans variety Harosoy 63 were supplied by the Michigan Foundation Seed Association Inc., East Lansing. The dry soybeans were soaked overnight in tap water at room temperature and the skin was removed by hand in running water. The soaking was proved to be essential by the fact that fermentation was about 100% successful, while unsoaked soybeans failed to ferment properly in 95% of the cases. The peeled soaked soybeans were boiled in a steam jacketed kettle for 30 minutes, cooled and drained prior to inoculation with Rhizopus oligosporus NRRL 2710, which has been kindly supplied by the Northern Regional Research Laboratory, Peoria, Illinois. One agar slant culture of mold suspended in about 3 ml of sterilized distilled water was sufficient to inoculate 300 g of cooked soybeans. The well mixed inoculated soybeans were then packed tightly into a stainless steel form, 25 x 10 x 3 cm, and wrapped with Dow Handiwrap plastic sheet perforated 30 31 with pinholes for air supply. For a smaller sample, the soybeans can be fermented in plastic bags. The tempe was fermented at 30-31°C in an incubator. The relative humidity was maintained at about 45% by placing water in the incubator. Samples for analyses were sliced from the fermenting tempe cake as desired. Sample preparation for electron microscopy Scanning electron micrOSOCOpy (SEM) Sliced, dried tempe about 1/4 cm. thick and 1 cm. in diameter was mounted on a special metal setting and gold coated. Raw tempe, similar in size to the dried one, was dipped in a 1% solution of 080 in phOSphate buffer for two 4 hours. It was then washed in solutions of increasing ethanol concentration, 25%, 50%, 75%, 85%, 95%, and twice at 100%, for half an hour in each solution. It was then dried in a Bomar SPC-900 EX critical point dryer and gold coated. Transmittance electron microscopy (TEM) Soybeans and tempe samples were cut into small strips about 0.5 x 0.5 x 2 mm. and placed in 5% glutar— aldehyde fixative in 0.1 M Sorensen buffer, and they were left in this solution for about 4 hours, and then washed 3 times (about 1 hour) with 0.1 M Sorensen buffer solution. 32 Subsequently they were fixed in 1% buffered OsO4 solution, embedded in Spurr's resin, sectioned in Porter—Blum MT2 ultramicrotome, stained in uranylacetate and lead-citrate, mounted and examined in a Philips 300 electron microscope. Extraction of lipids For comparison, two methods of lipid extraction were used. Method I About 100 g tempe was sliced thinly and dried in a 29 inch vacuum oven at 70 C for 5 hours. The vacuum-dried tempe was then ground into a fine powder (20 mesh). Fifteen grams of ground tempe was wrapped in Whatman filter paper and extracted with petroleum ether in a Soxhlet apparatus for 24 hours. The ether extract was filtered and washed with diethylether, dried under nitrogen, weighed and stored under nitrogen at 0°C for further analysis. Free fatty acids were separated according to Mattick and Lee (1959). After drying under nitrogen over sodium sulfate, the glycerides and free fatty acids were weighed. After apprOpriate esterification of the glycerides and the free fatty acids, each sample was injected in a Perkin Elmer 900 chromatograph, and SP 1000 liquid phase on Chromosorb was the packing material of the column. The temperature of the column was set at 200 C. 33 Method II About five grams tempe was blended in a Waring blender for 2 minutes with 100 m1 Chloroform : Methanol = 2:1 (v/v), according to the method of Folch g: gl. (1957). The finely blended homogenate was filtered through Buchner vacuum filter on Whatman No. 1 filter paper. The lipid solution was then dried in a flash evaporator at 40°C. After most of the chloroform methanol solvent was evaporated, the free fatty acids were separated from the glycerides, as in Method I. The glycerides and free fatty acids were then dried in a vacuum desiccator overnight and weighed. After ester- ification, each sample was injected in a Perkin Elmer 900 chromatograph, containing a DEGS column and programmed at loo-200°C. Separation of free fatty acids from glycerides The free fatty acids were extracted by the method of Mattick and Lee (1959). A mixture of standard free fatty acids (Applied Science Laboratories Inc.) was treated similarly. Figure 6 shows the steps in this extraction method. Diazomethane preparation and esterification of fatty acids Diazomethane was prepared in ether on the ice bath to a bright intensive yellow color as shown at Figure 7. 34 1 gram oil, dissolved in 17.5 ml diethylether 17.5 ml petroleum ether 6.5 m1 ethanol 95% add 12.5 ml of 1% Na2CO3 solution shake in separatory funnel for 30 seconds stand until two layers are formed \\ ethereal phase add 1.5 ml 95% ethanol 2 3 ethereal phase, add 1.5 ml 95% ethanol 2 3 ethereal phase, add &/’////,,////”’/I 6.5 m1 H20 0 l collection of lower aqueous phase i glycerides add 1.5 ml 10% H2804 add 12.5 ml solvent (diethylether:pet. ether=l:l) shake in separatory funnel and let stand I collect ether solution 4 repeat ether extraction 3x J! add 1 gram Na SO V 2 4 filter 1. dry under N W 2 free fatty acids ready for esterification Figure 6. Free fatty acid extraction (Mattick and Lee, 1959). aqueous phase 7.5 ml 1% Na CO sol 5.0 ml 1% Na CO sol 35 -—7 add 1 - 2 spatula N-methyl N'-nitro-N- __ _j nitroso Guanidine \ (MNNG) 97%. (Aldrich Chemical Company, Inc. :3\ Milwaukee, Wis. 5“ r-- 53233.) 0 ——-4 ° \ ° \ 0 ° —‘ O c a _, w N O : diethylether O anhydrous / O O O o c -_1. __- oO / diazomethane diethyl- ether 0 e -f O 0 ”"' 6 I’ll) 60% on y ' 2: ‘3‘ I f; 0 \ AI ice bath 1" 9.9.0.9.} Figure 7. Diazomethane preparation (Schlenk & Gellerman, 1960). 36 About 2 ml of diazomethane solution was added to the fatty acid sample, and a few m1 of 10% methanol solution in diethyl ether were added to speed up the reaction. Yellow color will persist when bubbling stOps in about 15 mins. at room temperature (25°C) if a sufficient amount of diazomethane was added. When the bubbling stOps, blow the excess diazomethane and ether under N then 2! dissolve the fatty acid methyl ester in diethyl ether to a desired volume (10 m1). A few 01 of the ester solution was ready for injection in the Gas Chromatograph. Transesterification of glycerides The glycerides remaining after the extraction of the free fatty acids were transesterified using the method of Mason and Waller (1964a) and Mason g: gl. (1964b). Approxi— mately 50 mg of glycerides were added to a reagent con- sisting of 3.5 ml dry benzene, 0.25 ml 2,2 dimethoxypropane (DMP) and 1.25 ml methanolic—HCl in a 15 m1 screw—cap vessel. After mixing to allow an even reaction, the mixture was left overnight, or at least for 6 hours, at room temperature (22°C) to ensure a complete transesterifi— cation. The entire sample was neutralized in 30 min. with one gram of neutralizing agent consisting of sodium bicarbonate, sodium carbonate and sodium sulfate anhydrous mixed in the proportions 2:1:2. This neutralizing agent was dried overnight before use at 110°C. 37 A few ul of this solution was injected into a gas chromatograph for fatty acid ester analysis. Pure glycerol was treated similarly and served as a standard elution peak as isopropylidineglycerol (Mason g3 g£., 1964a). Deep frying of tempe The tempe was cut to pieces 2.5 x 7 x 0.5 cm, weighed and deep—fried for 3—4 minutes in 150 m1 coconut oil in a 1000 ml glass beaker on a hot plate. The temperature of the oil was 180—210°C. The fried tempe was drained at room temperature, weighed and the oil was extracted and analyzed. Total plate count To follow the growth of bacterial population in each step of tempe preparation and fermentation, total plate count procedure was used. 50 g sample was blended in a sterile blender with 450 m1 of 0.1% peptone dilution blank for 1 minute. One ml of sample suspension was transferred into 99 m1 dilution blank, and repeated until the desired dilution 5-107). One ml of the diluted was obtained (about 10- sample was placed in a sterile petri dish, mixed with warm Bacto Plate Count Agar, and incubated at 30°C for 48 hours. To prevent the growth of mold from tempe, 100 ppm Actidione (Upjohn Company) was added to the agar medium. The bacterial count was computed per gram of sample. 38 TBA method for determination of oxidative stability Proposed TBA reaction (Sinnhuber g3 g£., 1958a; Dahle gggg£., 1962) N /N Hs—/ on 0 S = O l + ('3' HCl I N g — CH2 — a N \ / CH _. CH :- \ C\{ H O OH H/ 2 \OH TBA Malonaldehyde TBA chromogen SH H . + 2H 0 CH— 2 \\ OH Patton gg gi. (1951) identified the active color- producing compound in TBA (2 thiobarbituric acid) reaction with rancid oil was malonaldehyde. Sinnhuber g3 gi. (1957, 1958a, 1958b) suggested that two molecules of TBA condensed with one of malonaldehyde to produce the color substance. Some of the advantages of TBA test compared to other tests for rancidity are as follows: 1. The test can be performed on the whole food rather than on extracted fat. 2. Therefore the test can be expected to measure oxidation products of protein-bound lipids and phospholipids which would not be extracted by ordinary fat solvents. 39 3. The TBA test correlates better to the organoleptic rancidity testing (correlation coefficient, rS = 0.89) than the peroxide or carbonyl tests. One disadvantage is that malonaldehyde appears to contri- bute a relatively small fraction of the total odor complex of rancid sample, and by itself does not taste or smell bad (Tarladgis g3 g£., 1960; Sidwell g3 gl., 1954). The test procedure involved the heating of the food sample with a strong acid solution. This step appears to be essential for the liberation of malonaldehyde from some precursors as well as for the condensation of malonaldehyde with TBA. Tarladgis g3 gi. (1960) developed the distillation method for TBA test. They investigated the factors that might affect the maximum liberation of malonaldehyde: 1. pH of the material to be distilled 2. effect of time of heating during distillation 3. amount of distillate collected And they have concluded that maximum optical density was obtained at pH 1.5 for both standard solution of 1,1,3,3— tetraethoxyprOpane (TEP) which yields malonaldehyde on acid hydrolysis and cooked meat sample. The longer the time of heating during distillation will provide the increased color development on meat sample, but decreased on standard TEP. It was thought that fat oxidation may occur in meat Sanmfles during the prolonged distillation time. In the TEP Stfiindard the greatest amount of malonaldehyde was obtained When the distillate was collected in the shortest time 40 possible (i.e., the highest heating available). From 100 m1 of water blended sample, the maximum optical density was obtained when 50 ml of distillate was collected. The heating time of the distillate and TBA reagent mixture was 35 minutes in a boiling water bath. Distillate can be held over 24 hours at room temperature with very little change. However, after the reaction of the dis- tillate with TBA, the optical density will increase about 10%, if the solution is held for 24 hours. Reagents TBA Reagent: 0.02 M 2—thiobarbituric acid in 90% glacial acetic acid. Bring into solution by warming slightly in a boiling water bath. HCl Solution: 1 part of concentrated HCl to 2 parts of distilled water (approx. 4N). Procedure Blend a 3 g carefully weighed sample, with 50 ml distilled water in Waring Blender for 2 minutes. Transfer the mixture to a distillation flask (1000 m1) quantitatively by washing with an additional 48.5 ml distilled water. Add 1.5 m1 of HCl solution to bring the pH to 1.5. Add anti- foam agent and few boiling chips, assemble distillation apparatus with the highest heating available. In approxi— mately 10 minutes, 50 m1 of distillate was collected. Mix the distillate, filter, and pipette 5 m1 into 50 ml glass stOppered flask and add 5 ml of TBA reagent. Mix the content and immerse in a boiling water bath for 35 minutes. 41 A distilled water-TBA reagent blank should be pre- pared and treated like the samples. After heating, cool in tap water for 10 minutes, transfer a portion to a cuvette and read the Optical density of the sample against the blank at the wavelength of 528 nm. The Optical density is used as an arbitrary unit to compare the stages of rancidity of the samples. Sensory integpretation of TBA test value To make the values of the TBA test meaningful, the following sensory test was used for relating rancid flavor to TBA values. About 2 lbs. of dried soybeans were soaked over- night, peeled by hand and then boiled for 30 minutes. The boiled soybeans were blended in a Waring blender in about 2 liters of distilled water. The slurry was dried in oven at 200°F overnight on shallow pans. The dried mass was then ground in Arthur Thomas mill, 20 mesh screen. This soybean powder was used as a base for sensory evaluation. Rancid flavoring was prepared from diluted linoleic acid in ethanol by bubbling oxygen overnight at 100°C. The ethanol was then evaporated. Samples were prepared by mixing about 25 g soybean flour base with rancid linoleic acid in quantities from 0 to 1 ml in individual screw capped jars. The samples were presented to a test panel con— sisting of ten members. The instruction sheets, standard 42 rancid linoleic acid and soybean flour base were provided as reference. The values were tabulated, using arbitrary values as follows: - no rancidity = 0 - weak rancidity = 1 - strong rancidity = 2 Fractions of those numbers were used if a member's decision was in between those three categories. Phytic acid determination The phytic acid contents of soybeans and tempe were determined by using a combination of the methods of Makower (1970) and Wheeler and Ferrel (1971). l. Weigh accurately 2 g finely ground dry sample (Cyclone Sample Mill, about 40 mesh) into a 125 m1 Erlenmeyer flask. 2. Extract with 40 ml of 3% trichloroacetic acid (TCA, CCl COOH) for 45 minutes under mechanical shaking. 3 3. Centrifuge the suspension at 12,000 XLG for 10 minutes. 4. Transfer 10 ml aliquot of the supernatant into a centrifuge tube. 5. Add 5 m1 FeCl3 solution (containing 2 mg. ferric ion per ml in 3% TCA), to the aliquot by blowing rapidly from the pipet. 10. ll. 12. 13. 14. 15. 16. 43 Heat the tube and content in a boiling water bath for 1 hr. If the supernatant is not clear after 30 minutes add 1 or 2 drops of 3% sodium sulfate in 3% TCA and continue heating. Centrifuge for 10-15 minutes at 12,000 x G and carefully decant the clear supernatant. Wash precipitate twice by dispersing well in 20 ml 3% TCA, heating in a boiling water bath 5 to 10 minutes, and centrifuging. Repeat wash once with H20. Disperse precipitate in 5 m1 H20 and add 5 ml 0.6 N NaOH. Heat in boiling water for 45 minutes to coagulate Fe(OH)3. Centrifuge for 10—15 minutes at 12,000 x G and decant carefully. Wash precipitate with H2O, recentrifuge and decant. Precipitate was dissolved in 5 ml 0.5 N HCl with heating in boiling water for 10-15 min. until clear yellow color of FeCl was obtained. 3 Transfer to 100 m1 volumetric flask and make up to volume with 0.1 N HCl. Fe analysis Transfer 1 ml of solution from step 15 to a 25 ml volumetric flask, add 1 ml 10% hydroxylamine (NHZOH-HCI) solution, rotate flask and let stand few minutes. Add 9.5 ml 2M NaOAc solution and Reagent Reagent Pipette flask. 44 1 m1 O-phenanthroline solution (0.1 g/100 ml). Dilute with H20 to volume and mix. Let stand at least 5 min. and read at 510 nm against a water blank. Phosphorus determination (Murphy and Riley method, 1962) A: Dissolve 12 g of ammonium molybdate in 250 ml dis- tilled HZO' In 100 ml distilled H20 dissolve 0.2908 g of antimony potassium tartrate. Add both of dissolved reagents to 1000 m1 of 5N H SO (148 m1 concentrated H SO in 1 liter), mix 2 4 2 4 thoroughly and make to 2000 m1. Store in Pyrex glass bottle in a dark and cool compartment. B: Dissolve 1.056 g of ascorbic acid in 200 ml Of reagent A and mix. This reagent should be prepared as required as it does not keep for more than 24 hours. aliquot to be tested (0.5 ml) into 5 ml volumetric Add 0.8 ml reagent B, bring to volume with distilled H20. Read at 700 nm against a reagent blank. Protein evaluation methods Protein evaluation is most accurately carried out by feeding tests using the same animal species for which the protein is intended as a regular feed. Since this 45 seldom can be accomplished (especially for man), more convenient methods were developed using standard animal species (ig ziyg tests) and also purely chemical tests (£3 vitro tests). No single protein test is absolute since there are variations between tests. These variations arise from differences in strains or varieties, from variations in raw material sources and from changes induced by processing. Some of the frequently used standard animal tests are the following: 1. The Protein Efficiency Ratio (PER) is the gain in weight of a growing animal divided by its protein intake. Disadvantages of PER test are (a) PER is not a true efficiency ratio because not all the protein is used for growth, only that consumed above maintenance, and (b) varies with the acceptability of the feed. 2. The Biological Value (BV) is determined by nitrogen balance and is defined by the ratio of nitrogen retained/nitrogen absorbed. 3. The Net Protein Utilization (NPU) is the product of the coefficient of digestibility and the BV, and therefore represents the proportion of food nitrogen retained (nitrogen retained/nitrogen intake). The ig vitro tests were designed to save time, space and expense, some of which are described below: % essential amino acid = the PER larity. 46 The Chemical Score is determined by comparing the protein amino acid composition with that of a reference protein (whole egg protein). The chemical score is equal to the greatest percentage deficit in an essential amino acid of the protein being evaluated. Protein Score (FAQ/WHO procedure) is the ratio between the % limiting essential amino acid to the % corresponding amino acid of the reference pattern. essential amino-acid total essential amino acid x 100% Chemical methods for example: Urease Inactivation, Protein solubility test, Enzymatic and Microbio- logical tests. Procedure for the protein efficiengy ratio test In this present study of protein tempe evaluation test was used because of its simplicity and pOpu- This method was deveIOped by Osborne gg g;. (1917). There were four feed samples: High Protein Casein (Teklad Test Diets Company). Tempe, prepared as described in the beginning of this chapter, sliced, fried, most of the oil then extracted with hexane, and ground. Soybeans, prepared as the tempe but without fer- mentation. 47 4. Sesame seed supplemented tempe. Nine part of boiled soybeans were mixed with one part of ground sesame seed, fermented and treated as the tempe. The diets were prepared according to the AOAC method (Horwitz, 1970). The nitrogen content of the food was determined by the Micro Kjeldahl method, and the oil content by the Goldfisch diethyl ether extraction method. Protein = 6.25 x N. All feeds were standardized to meet the following composition: Protein 10% Oil 8% Salt mixture USP 5% Vitamin mix 1% Cellulose 1% Corn starch to make 100% Casein: High Protein Casein (Teklad Test Diets Company, 2826 Latham Dr., Madison, WI 53713). 911‘ Pure Corn Oil (Miesel Company, Detroit and Cleveland). BEAE‘ Salt Mixture, USP XVII (catalog no. 170890, Teklad Test Diets). Vitamin: Vitamin Mix AOAC, Teklad Test Diets, cat. no. 40055. 48 Cellulose: Non nutritive Fiber, cat. no. 160390, Teklad Test Diets. Corn Starch: Staley Company, Oak Brook, Ill. Experimental animal Twenty-four Spargue—Dawley male rats, were supplied by the Spartan ResearchAnimal Inc., Haslett, Michigan. Age: 21 days, weight: 51 g average. Acclimation period: 4 days on standard casein diet. At the end of acclimation period (beginning of test period), six rats were assigned into one group. Weight of rats in each group: 362.5 g, average weight of rat 60.4 g. Assay period The rats were kept in individual cages and both diet and water were provided gg_libitum. Body weight and food intake for each rat were recorded every 3 days. The experiment was terminated 28 days from the beginning of the assay period. Calculation Calculate average 28 day weight gain and protein intake per rat for each group. Calculate Protein Efficiency Ratio (weight gain/protein intake) for each group. Report quality sample ratio/casein ratio x 100%. RESULTS AND DISCUSSIONS Electron Micrographs of soybeans and tempe From the scanning electron micrograph (Figure 8) the pallisade-like cells of boiled soybean cotyledon measured about 25 x 1000. These large cells contain mostly the protein bodies (aleuron grains) and the spherosomes. The protein bodies (PB, Figure 9) which measure about 2 to 200 contain proteins, mostly glycinin, but also RNA, phytic acid and phospholipids. Neutral lipids are not an appreciable part of the protein bodies. The protein bodies are storage particles which dis- integrate on germination (Tombs, 1961, Wolf, 1970). The oil is located in the smaller structures called spherosomes (S, Figure 9) which are interspersed between protein bodies and are 0.2 to 0.5u in diameter. After soaking overnight, the protein bodies and the spherosomes were still confined in the cells and no indi- cations of breaking of the cell walls, Figure 9. However, after boiling the soaked bean for 30 minutes, the protein bodies and the Spherosomes were located in the cell wall areas (Figure 10). The disruption of the cell wall was 49 50 probably essential to facilitate the fast growth of the tempe mold. The photomicrographs of tempe (Figures 11A through e) show various stages of cell disruptions, from cell completely still intact, cell wall partially broken and some already completely lost their cell contents. The scanning electron micrographs (Figure 12 and 13) show that the RhiZOpus mold did not penetrate the soybeans more than two cell layers. Therefore, enzymes must be excreted by the mold deep into the bean mass to facilitate the chemical changes during tempe fermentation. An example of such excretion of enzyme is shown in Table 36. Chromatography of standard fatty acids Method I: Chromatographic profile of standard fatty acids mixture on SP 1000 column is presented in Figure 14. The conditions were set as the following: Column: length: 6 feet Liquid phase: SP1000 diam. : 1/8" wt : 10% Sample: 1% solution Support : Chromosorb W/AW standard ffa in mesh : 100/120 diethyl ether Carrier gas : N2 (Applied Science rotameter : 2.5 Laboratories Inc., inlet press : 50 Ann Arbor) Chart Speed : 30"/hr Size: 3.8 ul Detector : FID Flow rate H2: 24 air : 50 51 Figure 8. Scanning electron micrograph of boiled soybeans (magnification 200x). 52 ‘ kl... Cell wall U: 2 u 3 bodies Spherosomes in Prote C Transmittance electron micrograph of soaked soybeans. Figure 9. 53 ron micrograph of boiled tion 10,000 x). Transmittance elect soybeans (magnifica Figure 10. Figure 11. 54 Transmittance electron micrographs of tempe. 55 Magnification 3,200 x I 6.25u I 56 Magnification 160 x Scanning electron micrograph of dried tempe. Figure 12. Figure 13. 57 ISOuI Magnification 200 x Scanning electron micrograph of soybean tempe. 58 C16:0 o m a o o. m m u u 0 JJ 0 m 4.) 0 O C18:0 C18:1 I\ II‘ a“ I,‘ C18:2 l l H‘ C18:3 I" l ."—‘— ,u 1 l 1 ’l I l l I I ' 1 I l ' \ ' l ' | l . l l 1' 1 H 1 1 0 2 4 6 8 10 12 14 Elution time (min) Figure 14. The chromatographic profile of standard fatty acid mixture on a 10% SP 1000 column. 59 Temperature Range : 10x detect: 240°C Attenuation : 128x inject: 240°C column: 200°C Standard free fatty acids, 100 mg extracted as in procedure, esterified and diluted to 10 ml in diethyl ether. Sample injected: 3.8 ul = 38 pg. Atten: 128x. The height (h) and width (w, determined at half height) of each peak were measured and from the values of h x w, the prOportion of each peak which represented the individual fatty acid was determined. The results are as the followings: RggRg height RigER h x w R E3 cm cm C16:0 15.7 0.30 4.71 20.9 7.9 C18:0 8.4 0.55 4.62 20.5 7.8 C18:1 8.3 0.55 4.56 20.2 7.7 C18:2 6.4 0.70 4.48 19.9 7.6 C18:3 5.2 0.80 4.16 18.9 7.0 22.53 38.0 Response Factor C16:0 7.9/4.71 = 1.68 C18:0 7.8/4.62 = 1.69 Cl8zl 7.7/4.56 = 1.69 C18:2 : 7.6/4.48 = 1.69 C18:3 7.0/4.16 = 1.68 60 Because of the poor separation between stearic and oleic ac1ds (C18:0 and C18zl)’ these two peaks were determined as in Figure 14. the height and width of The proportion of fatty acids determined by this present method was very closely in agreement with the reference given by the company which supplied the standard fatty acid mixture. Method II: Chromatographic profile of standard fatty acid mixture on DEGS—PS column is presented in Figure 15. The conditions were set as the following: Column: Liquid phase : DEGS—PS length : 6 ft. wt. : 10% diam. : 1/8" Support : Supelcoport Sample: standard free fatty mesh : 80/100 acids, extracted, Carrier gas : N2 and dissolved in Rotameter : 2.5 diethyl ether (1% inlet press: 50 solution) Chart speed : 30"/h4 Temperature: Detector : FID detector: 240°C Flow rate H2 : 24 injector: 240°C air : 50 column : Range : 10x init: 100 Attenuation : 128x final: 200 rate: 10°/min 61 The separation between stearic and oleic acids was much better on the DEGS-PS column than on the previous SP 1000 column (Figure 15 and 14). The proportion of fatty acids in Method II was determined as in Method I, and similar results were obtained. Standard free fatty acid mixture, 100 mg was extracted as in the procedure esterified and diluted to 10 ml diethyl ether (1% solution). Sample injected : 2.2 01 = 22 u gram. Percentage calculation of each peak. RggRg height EEQEE. h x w R u gram C16:0 23.5 0.10 2.35 19.76 4.35 C18:O 14.0 0.18 2.45 20.61 4.53 C18:1 11.8 0.20 2.36 19.85 4.37 C18:2 9.45 0.25 2.36 19.85 4.37 C18:3 7.40 0.32 2.37 19.93 4.38 100.00 22.00 Response factor: C16:0 4.35/2.35 = 1.85 Cl8:0 4.53/2.45 = 1.85 C18:1 : 4.37/2.36 = 1.85 C18:2 : 4.37/2.36 = 1.85 C : 4.38/2.37 = 1.85 18:3 62 16:0 18:0 0 U) C: O 04 a"? “ C18:1 H O 4.) O 3 8 C18:2 18:3 . M. W, m 0 2 4 6 Elution time (min) Figure 15. The chromatographic profile of standard fatty acid mixture on a 10% DEGS—PS column. 63 Changes in fatty acid profile of tempe during fermentation The crude oil, free fatty acids and water contents of? tempe during fermentation are presented in Tables 9, 113a and 10b. Talole 9.—-Water and crude oil contents of soybeans and tempe. Sample Water (%) Crude oil (% dry basis) Boiled soybeans 67.0 31.87 18 run tempe 61.8 26.48 24 lhr. tempe 60.8 22.15 30 fir. tempe 62.8 23.40 40 fit. tempe 63.6 23.54 50 fur. tempe 63.2 25.41 70 fur. tempe 63.8 25.50 90 111:. tempe 63.6 22.38 112 rut. tempe 65.4 19.02 From the two fatty acid analysis methods, obviously M9t11C>C1 II was better than Method I in response sensitivity as VVEBJ.1 as the separation between fatty acids (Figure 14: 15 Eir1c1 Tables 10a and 10b). Based on the observations of Mnaad:ea (1965), Wagenknecht g3 gl. (1961) and Steinkraus mni c3<>~~workers (1960, 1961) that the amino acids composition ChaINQSECi only insignificantly and carbohydrates only changed in . . . . thea:Lr'solubility properties, therefore, the most IJ...--—__ 64 Table 10a.--Free fatty acids in soybeans and tempe, method I (mg in 100 g moisture free sample). sample C16:0 C18:0 C18:1 C18:2 C18:3 TOtal Raw soybeans 3.1 0.2 2.7 10.9 1.0 17.9 Boiled soybean 0.3 0.0 0.2 1.3 0.1 1.9 18 hr. tempe 187.5 37.0 371.8 641.2 119.7 1357.2 24 hr. tempe 322.3 74.3 754.9 1455.7 193.9 2801.1 30 hr. tempe 432.6 134.6 1346.0 1961.1 271.0 4145.3 65 hr. tempe 529.9 151.3 1512.5 1904.6 250.5 4348.8 72 hr. tempe 788.4 104.3 2427.8 3054.7 389.3 6764.5 96 hr. tempe 1078.3 277.2 2906.1 3335.6 494.6 8091.8 Table 10b.--Free fatty acids in soybeans and tempe, method II (mg in 100 g moisture free sample). sam‘fle C16:0 C18:0 C18:1 C18:2 C18:3 T°tal Raw soybean 12.98 2.10 10.49 31.48 5.56 62.61 Boiled soybean trace trace trace trace trace . . 18 hr- tempe 333.77 125.65 541.88 831.15 202.88 2,035.33 24 hr- tempe 433.67 232.60 988.88 1374.54 213.11 3,242.80 30 hr. tempe 725.89 257.89 1485.75 2057.14 302.70 4,829.35 40 hr- tempe 578.63 247.25 1449.45 1857.86 286.76 4,419.95 50 hr- tempe 463.48 234.02 1150.38 1787.39 265.65 3,900.92 70 hr- tempe 890.50 352.60 2284.70 3412.43 505.36 7,445.59 90 flr‘- tempe 1107.58 598.63 3344.40 4878.57 749.01 10,678.19 112 hr- tempe 901.45 531.16 2359.02 3988.44 451.50 8,231.57 \ 65 possible simple criteria of microbial or physiological activity of tempe fermentation was to observe the changes of free fatty acid being liberated and also the temperature. Tables 10a and 10b show that free fatty acids increased steadily during fermentation beginning from zero time to about 30 hours. After 30 hours the free fatty acid content of tempe remained essentially constant up to 50-65 hours. Beyond that time the free fatty acids increased again and deterioration signs appeared, i.e., loss of previously pleasant taste, ammonia smell, darkening of color, stickiness and general collapse of tempe structure. The free fatty acid continued to accumulate until the 90th hour; thereafter the tempe started drying and the free fatty acid content started to drop. Therefore essentially there were three phases in tempe fermentation: Phase I (0 - 30 hr.) : Rapid Phase Phase II (30- 60 hr.) : Transition Phase Phase III (beyond phase II) ': Deterioration Phase Based on the data of Tables 10a and 10b, Figure 16 was drawn which shows the three phases of tempe fermentation. The contaminating organisms which survive from bOiling process, mainly sporeforming Bacilli, multiply rapidly from 0 time to 19 hrs. with a generating time (g) of about 1.6 hrs. and then decline somewhat to a new eMporlential growth rate with a generating time of about 6- . . . . 7 hrs. (Figure 17) . The contam1nat1ng Ba01111 are shown in . . . the transmittance electron micrographs, Figure 18, from y 66 .Anoa pcm woa moanme Scum dunno cowumucwfiuou wcwusp couscoud mowom muumw mmum Houoe nun eon _ mfiwu coaumucoeumm om om on ow om ov om ON 0H _ _ 1 14 _ \ omosd c .umuofluouoo ommnd coeuwmcmna mound venom .ee «names ooo.m ooo.oa edmeq sex; eansrom 6 oor red spree Kine; ear; 6m 67 F“ wirflnvlflla . awn—Emu U\uG.DOO Hmuouo cedumucwfiww chL—é ¢CEQ+ cu. +5.5... Amp—59.3 05H» GOMumucmem on cm on ov om om a _ om Ffl‘l\E fr (1...... .OrF N 601 68 Figure 18. Transmittance electron micrograph of contaminating bacteria in tempe. 69 a 48 hr. tempe. Afterwards, repeated efforts to photograph this microorganism from tempe were unsuccessful. Most of the Bacilli were identified later as R. licheniformis and the rest were R. cereus (Stevenson, 1975). The fast growth rate of Bacilli coincided with the rapid phase of tempe fermentation (Figure 16 and 17) and with the slow phase of temperature increase (Figure 19), up to the 19th hour. The second half of the rapid phase from the 19th hour to the 30th hour, the temperature rose rapidly and at this time the Bacilli entered their new slower growth phase. The increase of temperature itself which never exceeded 45 C (Figure 19) was unlikely to cause the retardation of the growth of the thermoresistant spore- formers. One possibility cause of the slowering growth rate of the sporeformers was antibacterial compounds produced by the tempe mold especially during its peak of fermentation activity as already suggested by Wang gg g1. (1969). Up to the 30th hour of fermentation, the tempe mold was likely the dominant organism in splitting the soybean lipids to produce free fatty acids (Figure 16). At this time, the temperature of tempe started to fall (Figure 19), showing the declining activity of the mold; the content in free fatty acids leveled off and even decreased for some of them. .Uomm UN noumnsocfi am :« can ousumummsmu Econ um poumnsocw omfiou mo musumuwmfiou one .aa gunman A.muno mfiwu :oflumnsonH ov mm on mm on ma oH m 70 _ 4. _ . _ _ _ . ousuouomfiwu Eoou um pmumnsonH ousumuomfiwu uoumnsocH ow mm on mm ov mv (3°) aanezedmem 71 At the rapid and transition phases of tempe fer- mentation, apparently the Bacilli utilized the readily available nutrients produced by the mold. During the transition phase, both the dying mold and the Bacilli utilized the readily available nutrients remaining from the rapid phase. But after the transition phase was over, the Bacilli took over as the dominating organism in tempe, with a new outburst of fatty acid liberation. At this period, tempe deterioration commenced (deterioration phase). Tempe at the early stage of the deterioration period is still consumable, in small amount, as a flavoring, or in a Special recipe in Indonesia. The general trend of tempe metabolism activity in these present experiments was basically in agreement with the result of Wagenknecht g3 gi. (1961) which showed a stagnation trend of free fatty acid production at 25—30 hour of fermentation. To show the general trend of fatty acid composition changes during fermentation in the glyceride and free fatty acid portion of tempe, relative chromatographic analyses were made, and the results are presented in Tables 11a and 11b and summarized in Figure 20. The soybean oil fatty acid composition, as compiled by Herb and Martin (1970) from many oil analyses, is presented in Table 12. The results of the fatty acid analyses of soybean oils conducted in this work as presented in Tables 11a 72 Table lla.--Percent distribution of fatty acids among the free fatty acids (ffa) and the glycerides (gly) of tempe (from SP1000 column). Method I Sam le p C16:0 C18:0 C18:1 C18:2 18:3 TOtal Soybeans ffa 18.33 1.43 15.27- 58.86 6.11 100.0 gly 10.71 2.70 29.74 50.51 6.34 100.0 Boiled soybeans ffa 17.67 1.06 10.60 67.14 3.53 100.0 gly 10.04 3.29 24.65 54.63 7.39 100.0 18 hr. tempe ffa 13.81 2.73 27.40 47.24 8.82 100.0 gly 10.93 2.63 29.15 49.19 8.10 100.0 24 hr. tempe ffa 11.51 2.65 26.95 51.95 6.92 100.0 gly 10.50 2.53 27.40 51.80 7.77 100.0 30 hr. tempe ffa 10.44 3.25 32.47 47.30 6.54 100.0 gly 10.16 2.71 29.80 50.11 7.22 100.0 65 hr. tempe ffa 12.18 3.48 34.78 43.80 5.76 100.0 gly 12.18 1.98 32.94 45.00 7.90 100.0 72 hr. tempe ffa 11.65 1.54 35.89 45.16 5.76 100.0 gly 8.17 2.11 33.73 49.79 6.20 100.0 96 hr. tempe ffa 13.33 3.43 35.91 41.22 6.11 100.0 gly 9.70 2.24 33.65 48.59 5.82 100.0 73 Table 11b.--Percent distribution of fatty acids among the free fatty acids (ffa) and the glycerides (gly) of tempe (from DEGS-PS column). Method II sample C16:0 C18:0 C18:1 C18:2 18:3 T°t°l Raw soybeans ffa 20.90 3.35 16.72 50.17 8.86 100.0 gly 12.64 4.16 24.98 50.25 7.97 100.0 Boiled soybeans ffa 30.56 5.56 22.22 33.33 8.33 100.0 gly 11.92 4.29 24.79 49.82 9.18 100.0 18 hr. tempe ffa 16.40 6.18 26.64 40.79 9.99 100.0 gly 10.30 3.69 25.66 51.51 8.84 100.0 24 hr. tempe ffa 13.35 7.19 30.48 42.38 6.60 100.0 gly 8.09 3.43 32.47 49.39 6.62 100.0 30 hr. tempe ffa 15.03 5.33 30.77 42.60 6.27 100.0 gly 9.65 4.68 27.52 50.92 7.23 100.0 40 hr. tempe ffa 13.08 5.61 32.78 42.06 6.47 100.0 gly 8.78 3.38 30.49 49.66 7.69 100.0 50 hr. tempe ffa 11.88 6.02 29.49 45.81 6.80 100.0 gly 8.61 4.06 33.08 46.82 7.43 100.0 70 hr. tempe ffa 11.95 4.73 30.69 45.84 6.79 100.0 gly 8.23 4.41 29.58 50.24 7.54 100.0 90 hr. tempe ffa 10.37 5.61 31.31 45.69 7.02. 100.0 gly 9.37 4.21 30.02 49.71 6.69 100.0 112 hr. tempe ffa 10.96 6.46 28.65 48.45 5.48 100.0 gly 7.39 5.97 29.54 49.72 7.38 100.0 74 coflumucmfiuwm mcwuso .AQHH can mad moanma Eouu mauve mmEou mo mnofiuuom mowumoham can cwow muumu mmum mo Aw may COHuwmomfioo vwom auumm .om ousowm “any med» coflumucwEuwm OHH ooa om om on om om ov on on OH 0 d . q a H a . q . . u . o L. ‘1 K K L» \p\\0 ofiummuma aaaaaa n- x k K n ka : u|‘\‘wx£, P i - ......... o o o -----b----o:-.o--\- I llllllllllllllllllllllllllllllll Ibsucicilo‘ owcmaocwfl. 0 o IIIII lo.O|l llllllllllll nun uuuuuuuu o I y\ X a .r b.» ‘1. 1‘ 335mm 1 O .............. &. ............... In 0 O 0 :uaacl 'II’I 0 II / QJ&\L \\\ I \\ .\. om fl 8 4 K .\s.0\ cm 818...: x x ------..-------6-- lllOlluIItalIllutlIatllmttutud 10v 0 a II‘IIIIIIIIUO’I b '''''''''''''''''''' Ii‘"|'-'|||l‘lu‘l!nl|tlu-°‘\ " ’ UAW—www.mmwtlniau .. nnnnnn O 0!. O O III/I/ I 0 ll ’\ x w“ X [I] x < H 1!, X om onflumomam x vflom auuwu ooum O 8 (19:03 go %) spxoe K332; 75 Table 12.--Fatty acid composition of soybean oil (Herb and Martin, 1970). Mean values in % 0.09 9.27 4.25 47.65 35.78 2.97 and 11b, showed a higher prOportion in linolenic and linoleic acids but lower proportion in oleic and stearic acids compared to the values of Herb and Martin. The values corresponding to the two methods used (Tables 11a and 11b) were reasonably close. The data from those tables were summarized in Figure 20. Figure 20 shows the relative composition of fatty acids present as free fatty acids and glycerides in tempe during the course of fermentation. Oleic acid was liberated from the soybeans relatively more than other fatty acids. The percentage of palmitic acid, both in the frema fatty acid and the glyceride portions shows a decreasing trerui. This phenomenon was probably caused either by Palnnthic acid utilization by the mold or contaminating SPOI‘eformers or lost by other means. Linoleic and linolenic acids were liberated in a smalgleer degree than the other fatty acids and their per- centa-ges in the glyceride portion remained constant. This preservation of linoleic and linolenic in their glyceride forms probably contributes to the relative stability of tempe in storage, besides the antioxidant liberated during 76 the fermentation process. Linolenic acid especially in its free form, is known as the main cause of soybean instability during storage. The soybean oil is known as the most unstable of all vegetable oils due to its content of linoleic and linolenic acids. Soybean oil is more sensitive than other oils to iron, COpper and some extent of heating and aging (Cowan et 31., 1970; Dutton et 31., 1951; Antunes, 1971). Changes in the fatty acid composition of tempe after frying Tempe slices about 2.5 x 7 x 0.5 cm (weigh about 12-14 g) were fried in coconut oil at temperatures of 180 to 210°C for 3-4 minutes, and then cooled at room temper- ature in a vacuum desiccator for several hours. The oil was extracted according to the method of Folch et 31. (1957). Free fatty acids were extracted as described in the procedure. The analyses of fatty acids were carried out with GLC for both free fatty acid and glyceride fractions. In general, frying resulted in a decrease of free fatty acids, as shown in Table 13. Palmitic, stearic, oleic and linoleic decreased in all frying conditions, but linolenic increased after frying at 210°C for 3 minutes and at 190°C for 4 minutes. Heating and exchange of fatty acids between tempe and frying oil were considered as possible causes of the change in free fatty acids during frying. To investigate the heat factor similar sizes of 77 Table 13.-~Free fatty acids content of tempe before and after frying in coconut oil at 180°-210°C. mg of fatty acid per 100 g moisture free tempe Sample C16:0 C18:0 C18:1 C18:2 C18 3 TOtal Raw tempe 631.10 337.21 1396.79 1764.16 351.84 4129.26 Fried at 210°C for 3 minutes 355.96 267.47 917.84 1211.71 438.89 3191.87 .Fried at 190°C for 4 minutes 379.10 260.87 853.81 1197.57 419.53 3110.88 Fried at 180°C fk>r 4 minutes 185.81 152.02 552.84 573.83 253.37 1717.87 tempe were heated in oven at 200°C for 20 minutes, and the results are shown in Tables 14 and 15. Heat increased the free fatty acids in tempe, especially linolenic acid (Table 14) . TalajJe 14.—-Free fatty acids content of tempe before and after heating in an oven at 200°C for 20 minutes (g in 100 g sample dry basis). Sample C16-0 C18- C18'1 C18-2 C18-3 Total Raw tempe 0.579 0.289 1.423 1.663 0.226 4.181 Heated tempe 0.671 0.371 1.628 2.168 0.310 5.096 Increased (%) 15% 109 14% 30% 37% 22% \K 78 Table 15.--Fatty acid composition of free fatty acid and glyceride portions of tempe, before and after heating in an oven at 200°C for 20 minutes (in percent of total free fatty acid or glyceride). C . C 0 c . c . c . Total Raw tempe --free fatty acid 13.85 6.93 34.04 39.78 5.40 100.0 --glyceride 8.44 3.24 30.62 50.24 7.46 100.0 Heated tempe -—free fatty acid 13.17 6.24 31.94 42.55 6.10 100.0 —-glyceride 9.43 3.48 33.27 48.26 5.56 100.0 Linoleic and linolenic acids were liberated more, in relative terms, than the other fatty acids during heating. The percentages of linoleic and linolenic acids in the free fatty acid fraction increased from 39.78% and 5.10% to 42.55% and 6.10%, respectively. The consequence of these increases of free linoleic and linolenic acid was that the percentage of both acids decreased in the gly- ceride fraction from 50.24% and 7.46% to 48.26% and 5.56%, respectively after heating. To test the seepage of free fatty acids from tempe to the frying oil, ten grams of frying oil which has been used several times for frying tempe was analyzed. The free fatty acid composition extracted from frying oil is shown in Table 16. For reference, the fatty acid composition of fresh coconut oil was analyzed and appears in Table 17. From those two tables, it is obvious that fatty acid characteristic to coconut oil, caproic, caprylic, 79 w Aooav 0.0 mN.N om.b mm.m Hm.m Nv.ma mo.mv Hm.m mm.© mh.o Hmuoe U U U U U U U U U U .Awfiom mupmm Hmuou mo ucmoumm new Hflo uscoooo ammum CH mocflumowam wnu CH ucommmm Uflom wuumm mo Goeuflmomfiouul.ha magma w Aooav th.av Amm.HNv Amv.hav A©O.HHV Aom.mav Avm.oav Ah¢.mav AmH.HV AHV.HV Ao.ov mE om.av v5.0 v0.0 Hm.m v©.v m®.© «m.v ha.m om.o mm.o 0.0 Hmuoe m.mao N.mHU H.mHU o.wHU 0.0HU o.vHU O.NHU 0.0HU o.mU m.wo .AHHo scum OOH ca mac Hflo coummncs mnu CH mcflom wpumw mmum oc muoB muons .Amuson me moEHu Hmum>om mmfiou deflmum MOM poms coon mm: HHO mcflwum one .HHO mafimuw uncoooo ca pcsow mpflom >uumw mmumnu.oa magma 80 capric, lauric and myristic proportionately are much lower in free fatty acid portion of the frying oil, compared to their percentage in the fresh oil. On the other hand, the fatty acids characteristic to the soybean oil, palmitic, stearic oleic, linoleic and linolenic, proportionately were much higher in the free fatty acid portion of the coconut frying oil. This evidence led to a conclusion that free fatty acids seeped from tempe into the frying oil. The glyceride portion of coconut oil was also analyzed. Table 18 shows the fatty acid composition of coconut oil before and after frying. The compositions of fatty acids from those two oil samples were relatively unchanged. For reference, the mean value of Herb and Martin (1970) is presented in Table 19. In their publication, Herb and Martin compiled the results of fatty acid analyses of oils (same sample provided) from more than 25 labora- tories in the USA, Canada, and Sweden. Their conclusion was that the coconut oil was the most difficult to analyze (based on their standard deviation) compared to other oils analyzed. The result of the present study of fatty acid composition in coconut oil closely agrees with the values of Herb and Martin. The analysis of fatty acids in the glyceride portion of tempe before and after frying is presented in Table 20. The coconut frying oil apparently was absorbed by the tempe during frying as anticipated. The fatty acid 81 o.o NH.N an.m on.m vh.m Hm.oa mm.nv wm.m om.h mm.o .Aoema .cflpumz 6cm gummy Hflo useoooo mo :ofluflmomeoo anon spammuu.ma magma o.o v¢.H Hm.o mv.m mm.oa ww.na mo.mv vm.m om.m Hm.o o.o mn.a mm.m mm.m mm.m mm.ma ww.mv mm.m Hm.m mm.o Audos He deflwum Hmuwd o.o mm.m om.m mm.m Hm.m N¢.ma mo.mv Hm.m mo.m mn.o mcflmum mHOMmm mumao mumao Humau onmao onwao onwau oumao ouoao oumo oumo cam 0H0wwn .Hflo uncoooo mo coeuuom .mmeu mo mcflwum umumm moflumowam map mo COHuHmomEoo Uflom >0ummul.ma manme 82 00H mm.m mm.oN om.oH v>.m mm.w NH.HH mm.mm mm.m vm.m 0.0 mmudCHE v How UoomH um mmsmu UwHHm ooH mm.m Hm.om mm.mH mv.m vm.m vw.HH vm.mm Hm.m mm.m o.o mmudCHE v How . Uoomfl PM meou UmHHm OOH Hv.m mm.mH Ho.oH mm.m NH.m mo.HH mm.om vv.m H¢.m 0.0 mmusCHE m How UoOHm um omEou cmHHm OOH mm.© mo.Hm mo.mm mm.m mm.m 0.0 0.0 0.0 0.0 0.0 mmfiou 3mm Hmuoe m.mH m.mHU H.mHU o.mH o.oH o.vH o.mH 0.0H o.mo 0.0 onEmm .AUHom muumw m0 we HHO usCoooo CH mCHmHm Houmm CCM muommn .mmEmu mo CoHuuom mcHHmomHm mCu mo COHuHmomEoo UHom muummll.om mHnme 83 content of glyceride portion of fried tempe was a result of equilibrium of the two components, i.e., the soybean oil and the coconut oil. The fatty acids characteristic of coconut oil were found in fried tempe i.e., caprylic, capric, lauric and myristic acids. Two common fatty acids of tempe and coconut oil, the palmitic and stearic, their percentages were relatively unchanged from the raw and fried tempe. On the other hand, the percentage of fatty acids characteristic to the tempe oil; oleic, linoleic and linolenic acids were reduced appreciably. Table 18 shows that the fatty acid profile of the glyceride portion of the coconut frying oil was unchanged compared to the fresh oil control. These data suggested that no glyceride migration had occurred from the tempe to the frying oil. Rancidity test of tempe using the TBA method Rancid oil was distilled in water, and the dis- tillate was reacted with TBA reagent (details were described in the Materials and Methods) and the absorbance was read in different wavelengths using a Beckman DU spectrophotometer. The result is shown in Table 21. The wavelength which produced maximum light absorption was between 528—530 nm, which is the same with the wavelength determined by Tarladgis e3 31. (1960) and Sidwell gt 31. (1954). 84 Table 21.--Absorbance reading of rancid oil sample in different wavelengths. Wavelength (nm) Absorbance 518 0.70 520 0.78 522 0.84 524 0.88 526 0.92 528* 0.94 530* 0.94 532 0.92 534 0.86 536 0.78 538 0.69 540 0.59 * Maximum light absorption. The correlation between the TBA value (as absorbance) and rancidity, was determined by a ten-member panel. The result is presented in Table 22. Three panel members failed to detect any rancidity in sample 3, while only one failed to detect any rancidity in sample 4. All ten members recognized rancidity (by smelling) in samples 5 and 6. For practical purposes, it was therefore suggested to use the absorbance value of 0.300 as borderline indi- cator between weak, not readily detected rancid smell, and readily detected pronounced rancid smell in soybean flour. To compare the stability of soybeans and tempe to rancidity 85 Table 22.--Sensory test and TBA value. Sample no. Mean sensory value* TBA value (absorbance) 1 0.0 0.160 2 0.4 0.180 3 0.8 0.235 4 0.9 0.270 5 1.7 0.300 6 1.9 0.320 *Mean value from ten panel members: 0 = no ran- cidity; 1 = weak rancidity; 2 = strong rancidity. stored at room temperature (25°C), several soybean and tempe samples were prepared and stored in plastic bags, as summarized in Table 23. The samples were stored at room temperature in plastic bags not under vacuum. The TBA values of the samples were determined and are presented in Table 24. The reports of Sidwell 33 31. (1954) and Rhee and Watts (1966) indicated that the TBA test correlates better with the stability of an oil than do the peroxide or carbonyl tests. They showed that soybean oil had a higher TBA value than cottonseed oil even when they had approximately the same peroxide or carbonyl values. They concluded that the TBA test correlates better than the other tests with the unacceptability of oil flavor, and they suggested an association of the TBA value with the linoleic and linolenic acids content of a fat. 86 Table 23.-—Soybeans and tempe sample S. Sample no. Treatment 1 Soybeans soaked, peeled, vacuum dried and ground 2 Soybeans soaked, peeled, boiled 30 min. vacuum dried and ground 3 Soybeans soaked, peeled dried and fried in coco 200°C Tempe, vacuum dried and , boiled 30 min. vacuum nut oil for 1 min. at ground Tempe, freeze dried, and ground Tempe, dried and fried in coconut oil at 200°C Table 24.--The TBA value of samples described in the Table 23. Sample no. and time storage TBA value as absorbance 1 soybeans 1 month 2 soybeans 1 month boiled 3 soybeans 1 month boiled, fried 4 tempe, dried 2% months (3% months) 5 tempe, freeze 2% months dried 6 tempe, freid 1% month 7 raw untreated soybeans (ground) 8 fresh soaked, boiled, ground and dried soybeans > 2.00 0.165 0.360 0.165(0.170) 0.125 0.115 0.450 0.235 87 The fried soybeans showed a higher TBA value than the fried tempe after storage of more than one month (Samples 3 and 6). This observation closely agreed with the study conducted by Gyérgy gt gt. (1964) on the anti- oxidant trihydroxyisoflavone which was liberated during tempe fermentation from raw soybeans. This liberated antioxidant was possibly responsible for the protection of tempe from oxidation. In tempe, dried or fried (samples 4 and 6, Table 24) and stored at room temperature up to 2.5 months, showed relatively low TBA values. The tempe had no distinct ran— cid or beany flavor. On the other hand, using the TBA value of 0.300 as the border value for rancidity, the fried soybeans stored for one month (sample 3) was then cate— gorized as rancid. The dried soybeans (sample 1) also had a much higher TBA value than the boiled soybeans and tempe after storage. Freshly boiled and dried soybeans showed a lower TBA value than the freshly ground unboiled soybeans (samples 7 and 8). In this case, boiling and drying probably reduced the "beany flavor" which also reduced the TBA value. The boiling perhaps inactivated the lipoxy— genase in soybeans and prevented the soybeans from ran— cidity during storage. Rackis gt gt. (1970) investigated the flavor con- stituents of soybeans and soybean flakes and concluded that the TBA-reactive substances presumably were protein-bound 88 and associated with the non—lipid fractions. They also reported that there were no clear correlations between the TBA value and the organoleptic evaluation. In this present study, the TBA test appeared not Specific to the degree of fat oxidation or rancidity only, but also showed a positive correlation with raw soybean flavor. The raw soybeans showed a strong beany flavor in its distillate, but not rancid odor, and possessed a high TBA value of 0.450. This phenomenon had also been reported by Sessa gt gt. (1969). The TBA test of rancidity in soybeans and soybean products was therefore not a specific test because of the complication provided by the beany flavors. Phytic acid content in soybeans and tempe The following samples were prepared: 1. Soybeans, raw, broken and skinned. 2. Soybeans soaked overnight (300 g soybeans in 800 ml tap water) peeled and dried in vacuum at 87°C. 3. Soybeans soaked, boiled (30 minutes) and dried under vacuum at 87°C. 4. Tempe (fermented 30 hours), sliced and dried under vacuum at 87°C. All the samples were ground through the Cyclone Sample Mill into a fine powder. The phytic acid content was calculated from the absorbance reading as follows: 89 Mg. of phySic a01d A(510) per 100 g moisture = 0_783_ — 0.007 x dilution factor x free sample ’ 2.9546 Each sample was run in duplicate. The phytic acid contents of the samples are presented in Table 25. Table 25.-—Phytic acid content of soybeans and tempe. Sample Phytic acid (% moisture free) Sample I Sample II 1. soybeans, raw 1.40% 1.42% 1.42% 1.38% average 1.41% 1.40% 2. soybeans, soaked 1.39% 1.49% 1.35% 1.49% average 1.37% 1.49% 3. soybeans, boiled 1.07% 1.39% 1.02% 1.42% average 1.05% 1.40% 4. tempe 0.78% 1.11% 0.86% 1.11% average 0.82% 1.11% The presence of phytates (Figure 21) in foods may lead to deficiencies of iron, magnesium, zinc, calcium and possibly other elements in man and other non—ruminant animals (Berlyne gt gt., 1973; McBean and Speckman, 1974; Reinhold gt gt., 1973; Ranhotra, 1972; Ranhotra gt gt., 1974a). In ruminants the presence of active intestinal phytase makes phytate interference of mineral adsorption a much less serious problem (Anonymous, 1967; Pileggi gt gt., 90 0 3* HO 0 HO— —OH \ ¢ :0}! Z 0’P\ HO O H HO\ H H //P —o — H 0\ /OH O \ P::0 OH I H6 H0 ‘1?- OH ll O H Figure 21. Inositol hexaphosphate (phytic acid). Two possible structures. 91 1955; Maddaiah gt gt., 1963; Nelson gt gt., 1971; Ranhotra gt gt., 1974a). Some phytate hydrolysis occurs in man probably due to microbial phytases in the gut or to nonenzymatic cleavage (Nicolaysen and Njaa, 1951; Hegsted gt gt., 1954; Subrahmanyan gt gt., 1955). Because of the susceptibility of man to mineral deficiencies due to the ingestion of phytic acid, the reduction of phytic acid content of foods or balancing the diet with high mineral content is nutritionally advantage- ous. Therefore, the reduction of the phytic acid content during the fermentation of tempe can be considered as one contribution to the superiority of tempe as food compared to the unfermented soybeans. In the soaking water (300 gr soybean soaked over- night in 800 m1 tap water) there was 1.63 - 1.56 ppm phytic acid. This phytic acid probably came from the skin and the germ where the germination process is the most active. In the cotyledon the reduction of phytic acid during soaking and boiling was not significant. Mollgaard gt gt. (1946) indicated that most oilseeds, including soybeans, contain no phytase. Even if there was a weak phytase activity in soybeans, the boiling which preceeded the fermentation would inactivate it. The possibility of phytase activity from soybeans in tempe was zero. Therefore the phytase activity in tempe fermentation must originate in the Rhizopus mold. 92 The phytates yield inositol upon complete hydroly- sis. Inositol has nutritional properties as an anti- spectacled eye factor, antialOpecia factor and growth factor in rats as well as a growth factor for microorganisms. Since some phytate was hydrolysed during tempe fermentation, inositol may be formed and provided additional advantages to tempe as a food. Phosphorus determination The formula to calculate the amount of inorganic phOSphorus, based in Murphy and Riley's work (1962), is: A700 30.974 — x —————— (07016 0.213) 200 x dilution factor Phosphorus (Hg) = The inorganic phosphate contents in boiled soybeans and tempe were determined (in duplicate) and the results are presented in Table 26. Inorganic phosphorus balance calculation l Phytic Acid .>6 P + 1 inositol MW Phytic Acid : 660 MW P : 31 I Calculation was based on average value of: boiled soybeans phytic acid : 1.05% inorganic P : 15.4 mg/100 g. tempe phytic acid : 0.82% inorganic P : 82.2 mg/100 g In 100 g, the loss of phytic acid from boiling to fermented tempe is 1050 mg - 820 mg = 230 mg 93 Table 26.--Inorganic phosphate content of boiled soybeans and tempe (as p P per g moisture free sample). P Item Sample I Sample II Boiled soybean 122.63 142.63 186.74 219.00 average 154.68 180.82 Tempe 882.40 872.01 821.90 1020.85 average 822.15 946.43 Inorganic P gain = 82.2 mg - 15.4 mg = 66.8 mg. Theoretical gain of inorganic P (assuming only from phytic acid): %x6x31=65.1mg II Phytic Acid content of boiled soybean : 1.40% Phytic Acid content of tempe : 1.11% Inorganic phosphorus of boiled soybean: 18.0 mg/100 g Inorganic phosphorus of tempe : 94.6 mg/100 g In 100 g moisture free sample, the phytic acid loss during tempe fermentation was 1400 - 1110 mg = 290 mg or equivalent to 290/660 x 6 x 31 = 81.65 mg P. The actual P gain was 94.6 - 18.0 mg = 76.6 mg. According to these calculations the increase in P during the tempe fermenta- tion may be explained on the basis of phytic acid hydroly— sis. 94 Ranhotra gt gt. (1974b) investigated the total phosphorus and phytic acid phosphorus in several soyflour products, and concluded that all soybean flours tested were high in total P (about 600 mg/100 g fullfat soyflour) with most of it in the form of phytic acid. The phosphorous balance calculation was made on the assumption that inorganic P gain during fermentation was derived from the hydrolysis of phytic acid. In calculations from two experiments of soybeans before and after fermenta— tion, the actual increase in P was considerably close to the theoretical value from the phytic acid loss. The discrepancies of theoretical and actual values were due to analytical error, other sources of P rather than phytic acid, and also the possibility of chances of partial cleavage from the six P atoms present in one molecule of phytic acid. Investigation of phytase activity in tempe Crude enzyme extraction l. Blend 25 g tempe (fermented 30 hours) with 150 ml 2% solution of CaCl2 for about 4 min. intermittently. 2. Centrifuge the slurry at 12,000 x G for 15 minutes. 3. Filter through Whatman No. 1 in vacuum to remove traces of fat. 4. Fill in dialysis casing and dialyze in maleate buffer, pH 6.5 for about 30 hrs. 95 5. Centrifuge at 12,000 x G for 10 minutes. 6. Check the supernatant and precipitate for enzyme activity. Substrate preparation 2.5 x 10_3 M Na—phytate solution was prepared to give 1 x 10'3 M in 5 ml final solution of enzyme essay. M.W. Na—phytate = 923.8 Dissolve 1.15 g of Na—phytate in distilled water, bring the pH to 6.0 with dilute HCl and the volume to 500 ml with water. Buffer preparation Prepared according to Gomori (1955). Phytase essay The activity of the crude enzyme extract was assayed as shown in Table 27. Enzyme-substrate solutions were placed in test tubes with teflon caps and incubated in a water bath at 40°C. At the end of the incubation period, 1 ml of solution was added to 1 ml solution of 20% TCA to pre— cipitate all the protein. Centrifugation followed for 10 minutes. To 0.5 m1 of supernatant add 1.6 m1 Murphy and Riley Reagent B, bring to volume in 10 ml volumetric flask and let stand for over 10 minutes. Read absorbance at 700 nm against mixture of reagent and water blank. Calcu- late the phosphorous content using the standard formula 96 Table 27.--Enzyme activity assay in buffered substrate solution. Enzyme solution Substrate H20 Buffer* (ml) (ml) (ml) (ml) Supernatant l 0.1 2.0 0.9 2.0 2 0.5 2.0 0.5 2.0 3 1.0 2.0 0.0 2.0 4 2.0 1.0 1.0 2.0 i 5 2.0 0.0 1.0 2.0 ‘7' Precipitate** 6 1.0 2.0 0.0 2.0 *Acetate buffer 0.2 M pH 5.6. **Precipitate diluted in 10 ml water. for phosphorous. The phytase activity results are pre- sented in Table 28. Table 28 shows, solution No. 3 provided the most efficient enzyme activity in 2.5 hr. incubation time. The activity decreased at 24 hr. incubation. The precipitate of crude enzyme after dialysis also had a phytase activity, but the absolute quantity compared to the solution was small (tube No. 6, table above). The effect of pH on tgmpe phytase activity The enzyme was extracted from tempe and pre- cipitated with ammonium sulfate as described below: 97 Table 28.--Tota1 P liberated in ug, and enzyme activity as pg P produced/m1 enzyme/hour. 2.5 hr.incubation at 40°C Total P Activity 24 hr.incubation at 40°C Total P Activity Enzyme-Substrate Solution No. l 0.00 0.00 8.96 3.73 2 9.67 7.74 37.74 1.57 3 21.92 8.77 54.83 2.29 4 22.63 4.53 92.90 1.94 5 5.44 1.09 10.24 0.21 6 13.21 5.28 31.60 1.32 l. Blend 100 g tempe in 300 m1 CaCl 2% solution for 2 about 5 mins. 2. Centrifuge the slurry at 12,000 G for 15 min. 3. Separate the supernatant and filter through Whatman 44. 4. Add 200 g of (NH 80 to the filtrate. 4’2 4 5. Centrifuge at 12,000 G for 15 min. to precipitate protein. 6. Dissolve precipitate in 100 ml maleate buffer (pH 5.6). 7. Dialyze the solution in a cold room for 36 hours. 8. Centrifuge the solution at 12,000 G for 15 min. and discard the precipitate. 9. Store the supernatant containing enzymes at 35°F. 98 Phytase assgy The following enzyme-substrate solution was prepared: enzyme sol. substrate _3 buffer (Na-phytate 2.5x10 M) 1 m1 2 ml 2 ml The buffered enzyme-substrate solutions (pH 3.9 to 6.9) were incubated in 40°C water bath for 3 hours. The enzyme activity was determined as ug P/ml enzyme/hour and the result is tabulated in Table 29. Table 29.--Effect of pH on tempe phytase activity in acetate and maleate buffers. Activity pH* (ug P/ml enzyme/hr) % of max. activity 3.9 2.98 50.68 4.2 2.65 45.07 4.6 1.04 17.69 4.8 1.85 31.46 4.9 3.30 56.12 5.0 4.11 69.90 5.3 5.33 90.57 5.6 5.88 100.00 6.0 5.24 89.12 6.6 3.62 61.56 6.9 2.82 47.96 *pH 3.9 - 5.3 : acetate buffer pH 5.6 - 6.9 : maleate buffer 99 Table 29 shows that tempe phytase had a maximum activity at pH 5.6. The experiment was repeated by using a single buffer system of 0.1 M citrate throughout the entire pH range of 2.9 to 6.0. The total phosphate production is presented in Table 30 and the phytase activity in Table 31. Table 31 led to the conclusion that tempe phytase had a maximum activity at pH 5.6. Figure 22 was drawn based on the data in the Tables 29 and 31; this figure clearly shows the maximum 3 phytase activity at the pH of 5.6 regardless of the buffer system used. The partially purified tempe phytase in this present study apparently was not very different in pH activity from that of the phytase of germinated dwarfi beans, Phaseolus vulgaris, which had an optimum pH of 5.2 as reported by Gibbins and Norris (1962). On the other hand, the phytase extracted from animal small intestines had a higher pH optimum. Works of Bitar and Reinhold (1972) showed the pH optima for phytases extracted from the mucosae of the small intestines of rats, chickens, calves and human to be 7.0 — 7.5, 8.3, 8.6 and 7.4, respectively. Determination of Vmax and Km of tempe phytase The arrangement of buffered enzyme-substrate solutions are shown on Table 32. The final substrate concentrations were from 0 to 1.5 x 10—3 M. The 100 Table 30.--Tota1 production of P in pg per 5 m1 reaction mixture (10" M phytate, 1 m1 phytase extract in citrate buffers of various pH). pH 1 hr 2 hr 3 hr 2.9 0 0 5.887 3.1 0 0 4.947 3.4 11.851 29.251 39.747 3.7 16.084 42.889 59.499 3.8 19.187 44.300 59.499 4.0 19.376 46.651 58.559 4.2 20.316 46.651 56.677 4.4 23.608 46.651 57.618 4.6 25.312 46.651 54.796 4.8 27.370 51.354 59.499 5.0 30.850 55.116 72.667 5.2 38.657 61.700 74.548 5.3 45.523 64.522 92.419 5.4 52.107 64.522 114.992 5.6 62.641 81.452 130.982 5.8 61.701 72.046 125.339 6.0 63.582 72.516 119.695 101 Table 31.--Tempe phytase activity at different pH* levels and its percent of maximum activity. Activity pH (pg P/ml enzyme/hr) % of max. activity 2.9 1.962 4.00 3.1 1.649 3.36 3.4 13.242 27.02 3.7 19.121 39.02 3.8 20.390 41.60 4.0 20.740 42.32 4.2 20.845 42.53 4.4 22.047 44.99 4.6 22.301 45.50 4.8 24.293 49.57 5.0 27.543 56.20 5.2 31.452 64.18 5.3 36.197 73.86 5.4 40.900 83.45 5.6 49.009 100.00 5.8 46.501 94.88 6.0 46.579 95.04 *0.1 M citrate buffer. 102 .wmmumnm wmfiou How w>udo EdEHumo mm mm mumpwom mummHme z «.0 mumumom z m.o mumpuHo z H.o .ummmsm .NN wHDme ON CV 00 ow OOH ' XEW AQIAIQOE peAJesqo 103 Table 32.—~Buffered enzyme-substrate solutions. Substrate Buffer 2.5x10‘3M Citrate Substrate Tube Na-phytate Phytase pH 5.6 Water final no. (m1) (ml) (m1) (m1) conc. 1 0.0 1.0 2.0 2.0 0.00 2 0.2 1.0 2.0 1.8 0.10x10'3M 3 0.5 1.0 2.0 1.5 0.25x10‘3M 4 1.0 1.0 2.0 1.0 0.50x10’3M 5 1.5 1.0 2.0 0.5 0.75x10'3M fifi i 6 2.0 1.0 2.0 0.0 1.00x10‘3M 4 7 3.0 1.0 1.0 0.0 1.50x10'3M enzyme—substrate solutions were incubated in test tubes in a 45°C water bath, and P production was determined at 0, 10, 20, 30, 40, 50, and 60 minutes. The results are shown in Table 33. From the data of Table 33, the initial velocities of the phytate-phytase reaction at various substrate con- centrations were calculated using linear regression analysis and with the help of a Wang 600 desk calculator. These results are shown in Table 34. From Table 34, Figure 23 was drawn to show the correlation between sub- strate concentration, S, and velocity of reaction, V. To determine the Vmax and Km, a Lineweaver-Burk plot was prepared (Figure 24). The 1/V and l/S values are shown in Table 35. 104 Table 33.-~Effect of substrate (phytate) concentration on the P release by tempe phytase in pg/5 ml reaction mixture. substrate concentration 10- M time (min) 0.00 0.10 0.25 0.50 0.75 1.00 0 0.00 0.00 0 00 0.00 0.00 0.00 10 0.00 5.16 7.33 8.64 8.93 9.87 20 0.00 8.37 13.54 16.84 17.96 20.50 30 0.00 12.70 20.63 28.22 28.88 33.39 40 0.00 16.27 28.21 38.09 42.42 46.74 50 0.00 17.49 33.11 49.76 50.79 55.49 60 0.00 18.53 37.81 52.59 58.51 63.11 0.00 11.13 22.52 36.06 52.33 63.34 66.63 Table 34.--Effect of substrate concentration on the velocity of phytate-phytase reaction. initial substrate concentration, 10_3M 0.00 0.10 0.25 0.50 0.75 1.00 Velocity 0.00 3.77 6.62 9.38 10.28 10.98 (pg P/10 min.) Corre- . . 0.998 0.998 0.992 0.997 0.996 lation coef. (r) 0.989 105 .>UH>Huom omwu>sm omega C0 mCoHumuuCGOCoo mumuhsm mo uowwwm .mm wusmHm z muoH x .CoHumnquoCoo oumupmnsm m.H o.H m>.o m.o mN.o H.o u 2 TS x 3.0 n as l M 32.2 n 85> 11: (utm 01/5 60) KQIATQOE eseqqu 106 .CoHuome oumuanmlommuhnm mmewu 0:» new DOHQ xusmuuw>mmBoCHq 2\m¢H x m\H o.m _ 1 _ H _ _ Z_0IX(df>r1/um 01) A/I .vm musmmm 107 Table 35.—~Lineweaver—Burk plot of the phytate—phytase reaction. Values of 1/S and 1/V. s v 1/s 1/v 10_3M pg P/10 min 103/M 10 min/pg P 0.10 3.77 10.00 0.2655 0.25 6.62 4.00 0.1510 0.50 9.38 2.00 0.1066 0.75 10.28 1.33 0.0972 1.00 10.98 1.00 0.0910 4 I1 1.50 11.89 0.67 0.0841 3 The intercept with ordinate is 1/Vmax and the inter— cept with the abcisa is l/Km, and thence the Vm and Km ax can be determined. By regression analysis the following linear equation was obtained. 1/v = 0.0195 x 1/s + 0.0708 The intercept with ordinate is 0.0708, which is also l/Vmax, and therefore Vmax = 1/ (0.0708) = 14.1265 pg P/10 min. The intercept with abcisa is - 3.55 x 103/M which is also - l/Km, and _ -3 Km — 0.28 X 10 M 108 The work of Gibbins and Norris (1962) indicated the inhibition effect of substrate (Na—phytate) of 2 x 10-3M to the phytase activity of germinated dwarf bean, Phaseolus vulgaris. The Michaelis constant for this particular phytase was reported as 0.15 x 10_3M, a value not very different from that for the tempe phytase. Phytase activity in the tempe mold, Rhizopus oligosporus The Rhizopus grown on solid Potato Dextrose Agar (PDA) in two days at 30°C, was harvested and ground in the Polytron grinder in 2% CaCl solution and the enzyme 2 extraction was proceeded as in tempe enzyme extraction method. The enzyme activity was tested in buffer systems (citrate and maleate) of pH 2.9 — 6.3, with 1.0 x 10_3M Na-phytate as substrate; no activity was detected. Further studies were needed to elucidate the nature of activation or inhibition of this strong tempe phytase. The immediate practical questions is whether the production of phytase is activated by the presence of phytic acid in the substrate or whether any bacterial contaminants con- tribute to the production of phytase in tempe. The possibility of phytase activity naturally present in soybeans seems remote since after soaking over- night the beans was boiled for 30 minutes, and also a work of Mollgaard gt gt. (1946) indicated that most oilseeds including soybeans contain no naturally present phytase. 109 To investigate the possibility of substrate (Na- phytate) induction and bacterial contaminants interaction to the production of phytase, the following experiments were carried out: 1. Effect of phytate. 4.9 g PDA powder was mixed with 100 maleate buffer pH 6.3, stirred and heated. Then add 1.57 g of Na-phytate (acidified to pH 6.0 with 0.1N HCl) to the mixture (about 1.5% phytate) and was autoclaved in test tubes, at 250°F for 15 mins. Prepare as standard agar slants. 1 2. Sterile culture of Rhizopus in PDA-soybeans agar. Dissolve 5 g PDA in 100 ml water and add 5 g boiled soybeans, blended in a Waring blender. Then heated on hot plate, autoclaved in test tubes at 250°F for 15 minutes. Prepare as standard agar slants. Inoculate both groups of media asepticallyxflith 3. oligosporus, and incubated at 30°C for 2 days. Growth of Rhizopus on media Rhizopus grew poorly on phytate medium, growth only on the agar surface, very thin, stunted and produced a lot of spores. On the other hand, Rhizopus grew luxuriantly on soybean medium, thick and whitish in color due to lack of spores and filled the Spaces in test tubes. The mold grew on unsupplemented PDA less luxuriantly than on PDA—soybean, but much more mycelia were produced than on the PDA- phytate medium. After the mycelia were harvested, the 110 enzyme was extracted from the mycelia by grinding in a Polytron grinder with 2% CaCl2 solution. The phytase activity was assayed in maleate buffer pH 5.6 at 45°C with 1.0 x 10‘3M Na—phytate. Phytase activity The phytase activity of g. oligosporus is tabulated in Table 36. No phytase activity was detected in the PDA- phytate medium. On the other hand, in the mycelia harvested from PDA-soybeans medium a mild phytase activity was @ detected. After the mycelia were removed, including the rhizoid portion on the surface of medium, the clear medium was washed with water twice. Extract the agar medium with enzyme extraction procedure and phytase activity was assayed. A weak enzyme activity was detected in this medium, showing that the phytase was also excreted by the mold into the medium. From the above experiments of phytase activity of mold grown on PDA—soybeans and the lack of activity on PDA-phytate medium, it may be deduced that phytate did not induce the production of phytase in the tempe mold; apparently the protein in soybeans was involved in the production of the phytase. To test the validity of this hypothesis, PDA-peptone and PDA-peptone-phytate media were prepared. PDA—peptone-phytate medium 5 g PDA + 1 g peptone + 1.5 g Na-phytate (which was acidified to pH 6.0 with 0.1 N HCl), and were made into 111 Table 36.-—Phytase activities of Rhizopus oligosporus NRRL 2710 grown on PDA, PDA—phytate, PDA-soybeans, PDA-peptone-phytate, PDA-peptone and PDA- coconut liquid and of the mycelia—free medium. Buffer Phytase Enzyme Substrate pH 5.6 activity sol. (Na—phytate, (maleate, (pg P/10 (ml) m1) ml) min/ml) Mycelia on PDA l 2 2 0.0 PDA-phytate 2 2 0.0 PDA—soybeans l 2 2 3.23 PDA—peptone— phytate l 2 2 3.95 PDA-peptone 1 2 2 21.94 PDA-coconut liquid 1 2 2 1.21 Medium (mycelia free) PDA-Soybeans 1 2 2 1.21 100 ml with deionized water. from this medium. PDAfipeptone medium Agar slants were prepared 5 g PDA + 1 g peptone, the volume was brought to 100 ml with H20, sterilized and prepared as agar slants. In addition, a PDA—coconut liquid medium was also prepared to see whether this coconut liquid endosperm had the capacity to induce the production of phytase in Rhizopus. 112 PDA—coconut liquid medium Free coconut liquid was collected from a coconut purchased at a local store, and filtered. The pH was 5.6. Five grams of PDA was added to 100 ml coconut liquid, which was then sterilized and converted to agar slants. Rhizopus was inoculated on the media and incubated in 30°C for two days. The mold grew very well on PDA— peptone, and PDA-peptone—phytate, similar as on the PDA— soybeans medium. The mold grew less well on PDA-coconut liquid, and on the unsupplemented PDA medium. The phytase activities of mycelia grown on those media were assayed. Table 36 shows that phytase activity was the highest in the mycelia harvested from PDA-peptone medium. The activity was essentially similar from both PDA-soybeans and PDA—peptone—phytate, which suggested that the phytate naturally present in soybeans and Na—phytate artificially supplied did not promote the growth of mycelia. On the other hand, peptone promoted strongly both the growth of Rhizopus and the phytase activity of the mold. There was a seemingly positive correlation between the growth of mycelium and the phytase activity in the mold. Further studies are needed to quantitate this correlation using equivalent amount of mycelia. The coconut-liquid only supported minimally the production of phytase in the tempe mold. 113 The protein efficiency ratio values of soybeans, tempe and sesame— supplemented tempe Feed preparation: Feed code : 0 Type : Standard Casein diet Protein source : High Protein Case in (Cat. no. 160030, Tekland Test Diets Company) Analysis : Protein 87.0% Fat 1.2% Feed preparation: Feed to be prepared: 5000 g g Casein 574.7 (10% protein) Corn oil added 393.1 (8%) (400g-6.9g) Salt 250.0 (5%) Vitamin 50.0 (1%) Cellulose 50.0 (1%) Corn starch 3682.2 Total 5000.0 Feed code _: I Type : Fresh fried tempe Protein source: Soybean tempe (fermentation 30 hrs), fried in coconut oil at 200°C. for 3 minutes. Most of the oil was extracted with hexane in Waring blender and filtered. Tempe flour was dried in hood for overnight. 114 Analysis : Protein 55.65 % (Micro Kjeldahl AOAC) Oil 8.5 % (Goldfisch method) Feed preparation: Tempe flour Corn oil added (335.2 g — 64.0 9) Salt Vitamin Cellulose Corn starch Total Feed code : II Type : Fried boiled soybean g 753 (419 g protein = 10%) 271.2 (8%) 209.5 (5%) 41.9 (1%) {‘1' 41.9 (1%) “'1 2872.5 4190.0 Protein source : soaked and boiled soybeans prepared as for tempe but not fermented. The soybeans then fried in coconut oil at 200°C for 3 minutes. Most of the oil was then extracted with hexane as in feed I. Dried and ground into flour. Analysis : Protein 56.8% Oil 11.0% Feed preparation: Soybean flour Corn oil added (363.5-88.0 g) 9 800.0 (454.4 g protein = 10%) 275.5 (8%) 115 Salt 227.2 (5%) Vitamin 45.4 (1%) Cellulose 45.4 (1%) Corn starch 3150.5 Total 4544.0 Feed code : III Type : Sesame supplemented tempe Protein source : Soaked boiled soybeans prepared as feed I. Staley sesame was lightly ground in mortar to break the seeds. Protein analysis of boiled soybean was 17.2%, and sesame was also 17.2%. Nine part of boiled soybeans were mixed with one part of sesame (by weight). The soybeans—sesame mixture was inoculated with tempe mold and fermented for 30 hrs. Then fried in coconut oil at 200°C for 3 minutes. Extracted with hexane in Waring blender, dried and ground into flour. Analysis : Protein 57.21% Oil 10.16% Feed preparation: 9 Sesame tempe flour 725.0 (414.77 9 protein = 10%) Corn oil added 258.0 (8%) (33l.8-73.8) 116 Salt 207.5 (5%) Vitamin 41.5 (1%) Cellulose 41.5 (1%) Corn starch 2874.2 Total 4147.7 Test animal: 24 Sprague-Dawley male rats, supplied by the Spartan Research Animals, Inc., Haslett, Michigan 48840. Age: 21 days, weight 46—58 g. Acclimation period: 4 days with casein diet. Standard Diet (Feed Code 0) and drinking water gg libitum. After the acclimation period, the rats were assigned into four groups with six rats in each group. Each group of rats received one of four feed preparations. The average weight of rats in each group during 28 days of feeding experiment are summarized in Figure 25, and weight gains in each group of rats related with their age are shown in Figure 26. The rats in the casein diet gained more weight than the other three groups. The weight gains of all groups of rats dipped into a minimum when they reached 40-45 days of age. The Protein Efficiency Ratios were calculated for each protein and their variations were statistically analyzed. The PER of the proteins of casein, fried soybeans, fried tempe and fried sesame-supplemented tempe, in the 10% protein feed blends, for 28 days using male Sprague-Dawley rats was presented in Table 37. 1137 .msmo mN mchsc uch 0:» :H amououm Hugo» mcH an muqu Human omega pwucmEdedsmuoemmmm 6C0 mcmwnmom .odamu .CHommo 60m mmaoum anon no comm CH mumu no uCuHo3 omuuo>< .mN musmHm umou umccs mumHU gnu maHuMHuHCH Eouu mama mm ow vm NN om mH mH NH m w m 1 . . m d . H _ J . . wmemu < I mauwnaom a 1 cm mason wouC060Hmmsmuoammom o CHmmmo . < 1 “\\\o “\o\ O Q \ \ I Q\D\O O <\\u\\o\ 1 03 \ <\°\"\O Ga \ 1 \< \n\”u|l.\l|\\o\ a\ \o\\\.\. O L KOV\ l J\\\\\. : omH O 6 ‘dnoxb uses at austen 191 abeJeAe 118 .pmHm 0:0 cH :HmpOpC Hmuou WOH um .mumHe HCmummep pom .Amdouw\mumu my museum umu mo Ame CHmm qusz .mm musmHm mwmp .umu mo 06¢ mm mm me me me ov mm vm Hm mm _ H _ _ _ . _ . m . < u s\\\\ 1 /o mason o D I mCmon>0m < 1 mmEmu pmuCoEonmdmnmemmom D Cmemo . ’uteb lquSM 6 119 Table 37.——The PER values of casein, soybeans, tempe and sesame-supplemented tempe (values are mean of 6 male rats in each group; level of protein: 10%). Protein Gain in PER % Source of intake weight (28 Casein For casein protein (grams) (grams) days) PER PER = 2.50 Casein 28.93 98.00 3.39 100% 2.50 Soybeans 30.00 81.67 2.70 79.6 1.99 (fried) Tempe 26.92 68.17 2.53 74.9 1.87 (fried) Sesame- 30.65 85.00 2.77 82.0 2.04 supplemented tempe (fried) One-way analysis of variance and Tukey's test was used for statistical analysis. Total SS: 46 2 £2 Yij — (Y.. /24) 197.04 - 193.77 = 3.27 SS Y Treatment SS 4 ssT = (x y? /6) — (Y..2/24) . J. . i=1 = 196.33 - 193.77 = 2.56 Error SS: SS = SS - SS = 3.27 - 2.56 = 0.71 E Y T 120 Source of variation gt SS Mg Treatment 3 2.56 SST/3 = 2.56/3 = 0.85 Error 20 0.71 SSE/20 = 0.71/20 = 0.035 F0.05' 3'20M: 3.1 (table) Actual F = is: = 0.85/0.035 = 23.94 >> 3.1, there is E significant treatment mean difference, Specific test (Tukey's): F = (q0.05, 4,20) (JMSE76) = 3.958 x 0.0768 = 0.3040 The F of PER : casein and soybeans = 3.39 - 2.70 = values 0.69 casein and tempe = 3.39 - 2.53 = 0.86 casein and sesame supplemented tempe = 3.39 - 2.77 = 0.62 Because those Fvalues are bigger than 0.3040, so that there are significant differences between casein and all soybeans, tempe and sesame-supplemented tempe proteins. The F of PER : soybeans and tempe = 2.70 — 2.53 = 0.16 values soybeans and sesame - tempe = 2.70 - 2.77 = 0.07 tempe and sesame - tempe = 2.77 - 2.53 = 0.24 The Fvalues of soybeans, tempe and sesame-supplemented tempe are smaller than 0.3040, therefore no significant difference exists in the protein quality of soybeans, tempe and sesame-supplemented tempe. 121 The protein quality of tempe, sesame supplemented tempe and soybeans was statistically the same. The PER value of sesame-supplemented tempe was slightly higher than the unsupplemented tempe, but not significantly at the 5% statistical level. Figure 25 shows the rats receiving the sesame-supplemented tempe gained more weight compared to the unsupplemented tempe and the plain soybeans. This extra weight gain was due to extra consumption of feed and does not contribute to the improvement of sesame-supplemented tempe PER. Fried tempe protein was not significantly differ— ent, from the fried unfermented soybean protein. This finding was in agreement with the statement of Gy6rgy's (1961) that the heating of tempe at 66°C lowered the PER value in comparison with unfermented soybean flour. In the present experiment (Table 37), frying tempe at 200°C for 3 minutes resulted in a PER slightly lower than that of fried unfermented soybeans. The data from Chang and Murray (1949) also showed that autoclaving soybean curd lowered the PER to 1.59 from the PER of autoclaved soybeans of 1.73. Apparently, the curdling and the tempe fermentation processes modify the protein structure so that heat has more detrimental effect on the PER of the protein than on the plain soybean protein. Intact proteins are digested and absorbed in human subjects rapidly enough to be metabolized along with the free amino acids consumed at the same meal. In planning 122 dietaries, the administration of amino acids whether in the form of foods or crystalline amino acids can be applied satisfactorily with no significant difference (Clark, 1965). Consequently, the report by Van Buren gt gt. (1972) that at the end of tempe fermentation there was an accumulation of low molecular weight nitrogen compounds and solubilization of proteins, should not be expected in the improvement of protein quality in soybeans before and after fermentation. Furthermore, Murata (1965) had proved that protein content and amino acid profile of tempe and unfermented soybeans were essentially the same. Gy6rgy (1961) showed a significant PER improvement of lyophilized tempe over the unfermented soy flour using a lO-week feeding period in rats at the 10% protein level. However, drying tempe at 150°F (66°C) has resulted in similar PER value with similarly treated unfermented soy flour. Smith gt gt. (1964) also reported a slight decreasing PER value of tempe at 14% protein level from 2.42 to 2.35 after autoclaving the tempe for 40 minutes at atmospheric pressure. Processing has some detrimental effect on amino acid availability, especially if the in- takes were minimal. For example, 9% of the lysine in the wheat flour was destroyed or became unavailable during baking. Amino acids in foods high in carbohydrates are particularly susceptible to damage by heat (Clark, 1965). Results of Chang and Murray (1949) also suggested that autoclaved soybeans and unautoclaved soybean-curd have the 123 same PER value of 1.73. However, by autoclaving soybean- curd (110°C for 30 minutes) the PER drOpped to 1.59. Apparently soybean-curd protein was more sensitive to heat than the unprocessed soybeans. The biological values of many proteins are con- siderably changed by heat processing in the absence of discernible destruction of amino acids. The possible reasons for this change in biological value are as follows: 1. The digestibility of protein may be depressed. 2. The application of heat to a protein may promote certain combinations between terminal groupings that are resistant to proteolytic action (Sheffner, 1967). In this study, however, the fried tempe as well as the fried soybeans control were investigated, since dried raw tempe is never consumed in Indonesia. The supplementation of sesame seed in soybean tempe Numerous attempts have been made to improve the quality of protein sources for man be judicious supple- mentation with essential amino acids or by combining foods in a manner to achieve mutual supplementation. The objective of these supplementation efforts was to improve the balance or prOportions among the essential amino acids by overcoming a deficit without causing any amino acid to become either limiting or excessive. 124 Table 38 shows the amino acid compositions of tempe, the FAO reference pattern, egg, human milk, cow's milk, sesame seed and the pattern of amino acid require— ments of human beings. The reference pattern of amino acids was designed to evaluate the quality of protein foods, singly or in combination; it was developed in 1957 by the Committee on Protein Requirements of the Food and Agricultural Organi- zation (FAO, 1957). The assumption was made that the prOportions of individual amino acids in this pattern, were optimal. It has been demonstrated that efficient synthesis of tissue proteins occurs only when all essential amino acids are supplied simultaneously, and in prOper prOportions (Cannon gt gt., 1947; Geiger, 1947; Schaeffer and Geiger, 1947). Chang and Murray (1949) supplemented sesame seed to soybeans (dried autoclaved soybean: sesame - 10:4) and to soybean-curd (dried curd: sesame meal = 10:6). The PER of autoclaved soybean diet was 1.73 and increased to 2.17 when supplemented with sesame meal. The PER of standard whole milk powder was 2.30. Kies and Fox (1971) evaluated the protein nutritional value of TVP (textured vegetable protein), 1% DL-methionine enriched TVP and beef in human adults. From their experi- ments they concluded that at 8.8 g N intake level, TVP, methionine-fortified TVP and ground beef were of equal 125 .thd .Smmmo\0¢m .momH .wumuszW . .momH .mOCwHom mo hfiwcmod HmcoHumzN momH meHom CHououm m ooa\m Ho 2 a oH\m mm cummmumme H.m 6.6 m.m o.e e.o e.m m.m o.e m.m ¢.e m.v oaHHm> m.H o.H 6.H o.H o.H v.H v.H e.H m.H e.H v.H cmsmoumsue N.m o.m H.o m.o v.v s.v m.v e.¢ m.v o.m m.~ 0CHC00HC9 m.m m.HH . . . . . . . . . . o.oH m.m m.OH o.m oHumsoum Hmuoe o.e H.m . . . . . . m.m m.~ H.m m.m o.m m.~ 0chou>e m.H m.m m.v m.o 6.6 v.m m.w m.v v.4 m.m m.m 0CHC0HMHSC0CC v.0 o.m m.v . . N.w . . . . v.m m.v v.m N.v mcHOMIm Hmuoa H.m H.m . . . . . . o.H m.H m.o o.m m.m . . 0CHum>o m.H m.m . . m.m m.v m.m m.H m.m m.m H.m m.m 0CHCOHC002 m.¢ 6.8 m.m m.m 0.0 n.¢ >.m m.o m.m 0.6 m.¢ mCHosmHomH o.e H.o o.m m.oH m.OH m.e o.m o.OH H.m m.m m.g 0CHoswq H.m H.m e.oH m.e e.e H.m 6.6 m.e 6.6 6.6 ~.¢ 0CHm>C . . . . . . . . v.m e.~ e.~ e.~ N.N g.m . . 0CH6HumHm . . . . . . . . . . v.HH 0.5 e.m H.v 6.6 . . 0CHCH694 0H8: 0Hmsmm >H-~\H HN\Huo 600m xHHz xHHz cuwuumm muHsoa nHHno pummcH mmemmmm 400509 mm.3oo memesm mmmm moms N mucmEmHHsvmu pHom OCHEm mo Cumuumm H.60umHH OmHm mum C06 mo muCoEQHstwu cHom OCHEm 0C9 can mmEou .xHHE m.3oo .xHHE Cass: .mmmw mo ccm Cumuumm Cam :0 mo CoHuHmomEoo cHom OCHEM mCBna.mm mHnma . 036me 126 nutritional value, with a nitrogen balance of + 0.74, + 0.78 and + 0.72 respectively. On the other hand, at the 4.8 g daily N intake, ground beef was seemingly superior to both methionine - fortified TVP and the unenriched TVP. At 4.8 g N intake, methionine fortification at 1% level resulted in improvement in protein value but not to the extent of being equal to beef. At this low level N intake, the N balance of beef, methionine-enriched TVP and un- enriched TVP were -0.30, -0.45, and -0.70, respectively. Additional data from the same authors suggested that methionine supplemented TVP improved the PER value for growing rats, as shown in the respective PER values for TVP, methionine-enriched TVP and beef of 2.12, 2.82, and 2.37. Experimental results of the present study (Table 37) showed no improvement in the PER of fried, sesame- supplemented tempe over fried non-supplemented tempe. This non-significant improvement of sesame— supplemented tempe may be due to insufficient amount of sesame, as 9 parts of soy protein was supplemented with only 1 part of sesame protein. Another possibility is the fact that the deficiency of methionine in soybeans itself is not very great as in other beans, and the supplementa- tion effect is minimal. At 14% protein level, Smith gt gt. (1964) reported a PER improvement in freeze dried soybean-tempe supplemented 127 with 0.3% methionine. The PER value of methionine- supplemented tempe was 3.09 compared to 2.48 for unsupple— mented soybean tempe. The report of PER improvement from sesame- supplemented soybeans and soybean curd by Chang and Murray (1949) was probably due to a larger proportion of sesame used (10:4 and 10:6). Also they did not fry the soybean- sesame mixture in their experiment. Kies and Fox (1971) reported that fortification of p TVP with methionine had an insignificant effect on the N balance of adult males. The same authors, however, reported an increase in the PER of TVP supplement by methionine in rats and an improvement in the N balance of adolescent boys (Korslund gt gt., 1973). Russell gt gt. (1946) investigated the PER values of various legumes and the effect of methionine supple- mentation. For cooked and dried soybeans, the PER increased from 2.1 for basal diet to 2.8 for 0.1% methionine supplemented soybeans. The PER remained at 2.6 even though the methionine supplementation was increased to 0.6%. They noted that soybeans provided sufficient methionine to support growth in the basal diet, and the supplementation with methionine did not result in increases of PER as high as those observed with other legumes. SUMMARY AND CONCLUSIONS Tempe was prepared from soybeans and a pure culture of RhizOpus oligosporus. Electron micrographs of soybeans and tempe showed that boiling of soybeans resulted in the exit of protein bodies and spherosomes from the cell through the disrupted cell walls. The cell disruption apparently facilitates the tempe mold to grow. The mold penetrated only about two cell layers of soybean cotyledons. Based on the liberation of free fatty acids, microbial growth and organoleptic changes, the tempe fer- mentation may be differentiated into three phases: (1) rapid phase (0-30 hr), (2) transition phase, and (3) deterioration phase (beyond 60 hr of fermentation). It is believed that the contaminating, sporeforming Bacilli were mainly responsible for the deterioration of tempe. During tempe fermentation, oleic acid was liberated from glycerides more than any other fatty acid in the soybeans. In general, frying of tempe decreased its content in free fatty acids, except linolenic acid. This decrease was due to migration of free fatty acids from the tempe to 128 129 the frying oil. Linolenic acid also migrated, but an almost equal quantity of it was liberated from the tempe glycerides, and therefore its concentration in the tempe slightly increased after frying. No migration of gly— cerides from tempe to frying oil was detected. The TBA test was not specific for rancidity determination in soybeans since fresh, raw soybeans with a strong beany flavor, but not rancid flavor, had a high TBA value. Tempe, as well as boiled soybeans had lower TBA values than plain soybeans after storage at room temperature for at least one month. The phytic acid content of soybeans decreased from about 1.4% to 1.0% after tempe fermentation. The phosphate liberated was equivalent to the loss of phytic acid. There was a strong phytase activity in tempe. The phytase was produced by the tempe mold, R. oligosporus. Natural or added phytates did not promote the phytase synthesis by the Rhizopus; on the contrary it had an inhibiting effect. The tempe phytase had an optimum activity (sodium—phytate as substrate) at pH 5.6, and a Michaelis constant (Km) of 0.28 x lO-3M. The Protein Efficiency Ratio values of fried tempe (2.53) sesame-supplemented tempe (2.77) and plain soybeans (2.70) were not statistically different at the 10% total protein content of the diet. BIBLIOGRAPHY BIBLIOGRAPHY Anonymous. 1967. Effect of phytate on iron absorption. Nutr. Rev. 25:218. Anonymous. 1972. Vitamin 85 deficiency and brin lipids in the suckling rats. Nutr. Rev. 8:183. Antunes, A. J. 1971. Efeito dos Acidos Graxos Livres Sobre A Estabilidade, Paladar e Ponto de Fumaca de Oleos e Gorduras Comestiveis. (Effect of Free Fatty Acids on the Stability, Taste and Smoke Point of Edible Fats and Oils.) Thesis, State University of Campinas, Faculty of Food Technology, Brazil. Autret, M., and A. G- Van Veen. 1955. Possible sources of proteins for child feeding in underdeveloped countries. Am. J. Clin. Nutr. 3:234. Berlyne, G. M.; J. B. Ari; E. Nord; and R. Shaninkin. 1973. Bedouin Osteomalacia due to Calcium Deprivation caused by high Phytic Acid content of Unleavened Bread. Am. J. Clin. Nutr. 26:910. Bitar, K., and J. G. Reinhold. 1972. Phytase and alkaline phosphatase activities in intestinal mucosae of rat, chicken, calf and man. Biochimica et Biophysica Acta. 268:442. .Bodijn, K. B. 1958. Sydowia, Annales MyCOIOgiCi. Horn, Austria. (Jannon, P. R.; C. H. Steffee; L. J. Frazier; D. A. Rowley; and R. C. Stepto. 1947. Federation Proc. 6:390. Chang, I. C. L., and H. C. Murray. 1949. Biological value of the protein and the mineral, vitamin and amino acid content of soymilk and curd. Cereal Chemistry 26:297. Cheek, D. B. 1968. Human growth, growth, energy and intelligence. Lea and Febiger Publisher. 130 131 Clark, H. E. 1965. Utilization of essential amino acids by man. In: Newer Methods of Nutritional Bio— chemistry, Edited by A. A. Albanese, Academic Press. Clark, C. 1972. Calories and proteins in Indonesia. Bulletin of Indonesian Economic Studies 2:98-103. Cowan, J. C. 1969. Soybeans. Encyclopedia of Chemical Technology, Vol. 18, John Wiley and Sons Inc. Cowan, J. C.; C. D. Evans; H. A. Moser; G. R. List; S. Koritala; K. J. Moulton; and H. J. Dutton. 1970. Flavor evaluation of c0pper-hydrogenated soybean oil. J. Amer. Oil Chem. Soc. 47:470. Dahle, L. K.; E. G. Hill; and R. T. Holman. 1962. The Thiobarbituric acid reaction and the autoxidation of polyunsaturated fatty acid methyl esters. Arch. Biochem. and Biophys. 98:253. De Bruijn, J. D.; D: J. Frost; D. H. Nugteren; A. Gaudemer; G. W. M. LJ-meach; H. C. Cox; and W. Berends. 1973. The structure of Bongkrekic Acid. Tetrahedron 29:1541. Dutton, H. J.; C. R. Lancaster; C. D. Evans; and J. C. Cowan. 1951. J. Amer. Oil Chem. Soc. 28:115. FAO. 1957. Protein Requirements, Report of the FAO Committee, Rome 1955. Food and Agriculture Organization U.N., FAO Nutritional Studies 16. FAQ. 1971. Food Balance Sheet 1964-1966, Average, Rome. FAO/US HEW. 1972. Food Composition Table for Use in East Asia. Folch, J.; M. Lees; and G. H. S. Stanley. 1957. A simple method for the isolation and purification of total lipides from animal tissues. J. Biol. Chem. 226:497. Phaazier, W. C. 1967. Food Microbiology. McGraw-Hill Book Co. Geiger, E. 1947. Experiments with delayed supplementation of incomplete amino acid mixtures. J. Nutr. 34:97. GiIDIJins, L. N., and F. W. Norris. 1962. Phytase and acid phosphatase in the dwarf bean, Phaseolus vulgaris. The Biochemical Journal 86:67. 132 Gomori, G. 1955. Preparation of buffers for use in enzyme studies. In: Methods in enzymology I: 138-146. Edited by Colowick, S. P. and Kaplan, N. 0. New York Academic Press. Gydrgy, P. 1961. The nutritive value of Tempeh. Publi- cation 843:281-289. National Academy of Sciences- National Research Council, Washington, D.C. Gydrgy, P.; K. Murata; and H. Ikehata. 1964. Antioxidant isolated from fermented soybeans (Tempeh). Nature 203:870. Hegsted, D. M.; M. F. Trulson; and F. J. Stare. 1954. Role of wheat and wheat products in human nutrition. Physiol. Rev. 34:221. Herb, S. F., and V. G. Martin. 1970. How good are analyses of oils by GLC? J. Amer. Oil. Chem. Soc. 47:415. Hesseltine, C. W.; R. de Camargo; and J. J. Rackis. 1963. A mould inhibitor in soybeans. Nature 200:1226. Hesseltine, C. W. 1965. A Millenium of Fungi, Food and Fermentation. Mycologia 57:149. Hesseltine, C. W.; M. Smith; and H. L. Wang. 1967. New fermented cereal products. Development in Industrial Mycology 8:179-185. Am. Inst. Biol. Sci., Washington, D.C. Horwitz, W. 1970. Official methods of analysis of the Association of Official Analytical Chemists. AOAC, PO Box 540, Benjamin Franklin Station, Washington, D.C. 20044. Hudson, B. J. F. 1974. Spotlight on Lipids. Chemistry and Industry, June 15. .Jansen, B. C. P., and W. F. Donath. 1924. Mededelingen Burgelijken Geneeskundigen Dienst Ned. Indie, 1924:83. Kixes, C., and H. M. Fox. 1971. Comparison of the protein nutritional value of TVP, methionine enriched TVP and beef at two levels of intake for human adults. J. Food Sci. 36:841. KOJTEilund, M.; C. Kies; and H. M. Fox. 1973. Comparison of the protein nutritional value of TVP, methionine- enriched TVP and beef for adolescent boys. J. Food Sci. 38:637. 133 Latham, M. C.; R. B. McGandy; M. B. McCann; and F. J. Stare. 1972. Scope manual on nutrition. The Upjohn Company, Kalamazoo, Michigan. Maddaiah, V. T.; B. J. Hulett; A. A. Kurnick; and B. L. Reid. 1963. Availability of calcium phytate, sodium phytate and dicalcium-phosphate in chickens, hens and rats. Poultry Sci. 42:1286. Makower, R. U. 1970. Extraction and determination of phytic acid in beans (Phaseolus vulgaris). Cereal Chemistry, 47:288. Martinelli, A.; A. Filho; and C. W. Hesseltine. 1964. Tempeh fermentation: Package and tray fermentations. Food Technology 18:167. Mason, M. E., and G. R. Waller. 1964a. Dimethoxypropane induced transesterification of fats and oils in preparation of methyl ester for gas chromatographic analysis. Anal. Chem. 3:583. Mason, M. E.; M. E. Eager; and G. R. Waller. 1964b. A procedure for the simultaneous quantitative deter- mination of glycerol and fatty acid contents of fats and oils. Anal. Chem. 3:587. Mattick, L. R., and F. A. Lee. 1959. A note on a method for the extraction of free fatty acids from lipid material. Food Research 24:451. McBean, L. D., and E. Speckman. 1974. A recognition of the interrelationship of calcium with various dietary components. Am. J. Clin. Nutr. 27:603. Mollgaard, H.; K. Lorenzen; I. G. Hansen; and P. E. Christensen. 1946. On phytic acid, its importance in metabolism and its enzymatic cleavage in bread supplemented with calcium. Biochem. J. 40:589. Muralt, A. V. 1972. The developing brain and the damage inflicted by malnutrition: An introduction. In: Lipids, Malnutrition and the Developing Brain. Elsevier, North Holland. Murata, K. 1965. Nutritional value of Tempeh. Annual Report of Research Conducted Under Grants Authorized by PL 480. Osaka City University. Murphy, J., and J. P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chem. Acta 27:31. 134 National Academy of Science. 1963. Natl. Research Council, Committee on Protein Malnutrition, Publ. 1100. Nelson, T. S.; T. R. Shieh; R. J. Wodzinski; and J. H. Ware. 1971. Effect of supplemental phytase on the utilization of phytate phosphorous by chicks. J. Nutr. 101:1289. Nicolaysen, R., and L. R. Njaa. 1951. Investigations on the effect of phytic acid on the absorption of calcium in rats, pigs and man. Acta. Physiol. Scand. 22:246. O'Neal, R. M.; G. W. Pla; M. R. S. FOX; F. S. Gibson; and B. E. Fry. 1970. Effect of zinc deficiency and restricted feeding on protein and ribonucleic acid metabolism of rat brain. J. Nutr. 5:491. Osborne, T. B.; L. B. Mendel; and D. L. Ferry. 1919. J. Biol. Chem. 37:223. Patton, 8.; M. Keeney; and G. W. Kurtz. 1951. Compound producing the Kreis color reaction with particular reference to oxidized milk fat. J. Am. Oil Chem. Soc. 28:391-393. Pileggi, V. J.; H. F. deLuca; and H. Steenbock. 1955. The role of vitamin D and intestinal phytase in the prevention of rickets in rats on cereal diets. Arch. Biochem. BiOphys. 58:194. Rackis, J. J.; D. H. Honig; D. J. Sessa; and F. R. Stegerda. 1970. Flavor and flatulence factors in soybean protein products. J. Agr. Food Chem. 18:977. Ranhotra, G. S. 1972. Hydrolysis during breadmaking of phytic acid in wheat protein concentrate. J. Food Sci. 37:12. Ranhotra, G. S.; R. J. Lowe; and L. V. Puyat. 1974a. Effect of dietary phytic acid on the availability of iron and phosphorous. Cereal Chem. 51:323. Ranhotra, G. S.; R. J. Loewe; and L. V. Puyat. 1974b. Phytic acid in soy and its hydrolysis during breakmaking. J. Food Sci. 39:1023. Reinhold, J. G.; A. Lahimgarzadeh; K. Nasr; and H. Hadayati. 1973. Effect of purified phytates and phytate-rich bread upon metabolism of zinc, calcium, phosphorous and nitrogen in man. The Lancet 1:283. 135 Rhee, K. S., and B. M. Watts. 1966. Evaluation of lipid oxidation in plant tissues. J. Food Sci. 31:664. Roelofsen, P. A., and A. Talens. 1964. Changes in some B vitamins during molding of soybeans by Rhizopus oryzae in the production of Tempeh Kedelee. J. Food Sci. 29:224. Russell, W. C.; M. W. Taylor; T. G. Mehrhof; and R. R. Hirsch. 1946. The nutritive value of the protein of varieties of legumes and the effect of methionine supplementation. J. Nutr. 32:313. Schaeffer, A. J., and E. Geiger. 1947. Proc. Soc. Exptl. Biol. Med. 66:390. Schlenk, H., and J. L. Gellerman. 1960. Esterification of fatty acids with diazomethane on small scale. Anal. Chem. 41:173. Scott, W. O., and S. R. Aldrich. 1970. Modern soybean production. S and A Publications, P.O. Box 2660, Station A, Champaign, I11. Sessa, D. J.; D. H. Honig; and J. J. Rackis. 1969. Lipid oxidation in full fat and defatted soybean flakes as related to soybean flavor. Cereal Chem. 46:675. Sheffner, A. L. 1967. In vitro protein evaluation. In: Newer methods of Nutritional Biochemistry, VIII. Edited by A. A. Albanese, Academic Press. Sidwell, C. G.; H. Salwin,; M. Benca; and J. H. Mitchell, Jr. 1954. The use of thiobarbituric acid as a measure of fat oxidation. J. Am. Oil Chem. Soc. 31:603. Sinnhuber, R. O., and T. C. Yu. 1957. 2-Thiobarbituric acid method for the measurement of rancidity in fishery products. Food Technology 11:104. Sinnhuber, R. 0.; T. C. Yu; and T. C. Yu. 1958a. Charac- terization of the redpigment formed in the 2- thiobarbituric acid determination of oxidative rancidity. Food Research 23:626. Sinnhuber, R. O., and T. C. Yu. 1958b. 2-Thiobarbituric acid method for the measurement of rancidity in fishery products. II. The quantitative deter- mination of malonaldehyde. Food Technology 12: 9-12. 136 Smith, A. K.; J. J. Rackis; C. W. Hesseltine; M. Smith; D. J. Robbins; and A. N. Booth. 1964. Tempeh: Nutritive value in relation to processing. Cereal Chemistry 32:1412. Spaeth, R. S. 1974. Part III. Early soy protein research awaits a good economic climate for growth. Food Product Development 5:41. Stahel, G. 1946. Foods from fermented soybeans--as prepared in the Netherlands-Indies. J. N. Y. Botan. Gard. 47:261 and 285. Steinkraus, K. H.; Y. B. Hwa; J. P. Van Buren, M. I. Provvidenti; and D. B. Hand. 1960. Studies on Tempeh--An Indonesian fermented soybean food. Food Research 25:777. Steinkraus, K. H.; J. P. Van Buren; and D. B. Hand. 1961. Studies on Tempeh—-An Indonesian fermented soybean food. Publication 843:275-279, National Academy of Sciences, National Research Council, Washington, D.C. Steinkraus, K. H.; C. Y. Lee; and P. A. Buck. 1965a. Soybean fermentation by the Ontjom Mold Neurospora. Food Technology 19:119. Steinkraus, K. H.; J. P. Van Buren; L. P. Hackler; and D. B. Hand. 1965b. A pilot plant process for the production of dehydrated Tempeh. Food Technology 19:63. Stevenson, K. E. 1975. Personal communication, Michigan State University. Subrahmanyan, V.; M. Narayanarao; G. Ramarao; and M. Swaminathan. 1955. The metabolism of nitrogen, calcium and phosphorus in human adults on a poor vegetarian diet containing Ragi (Eleusine coracana). Brit. J. Nutr. 9:350. Tarladgis, B. G.; B. M. Watts; and M. T. Younathan. 1960. A distillation method for the quantitative deter— mination of malonaldehyde in rancid foods. J. Amer. Oil Chem. Soc. 37:44. Tombs, M. P. 1961. Protein bodies of the soybean. Plant Physiol. 42:797. Toyo Spinning Co. 1974. Soybean flavor improvement. Japanese Patent 4 908 264. Food Sci. and Techn. Abstr. 6:9J12l4. 137 United Nations. 1974. Statistical Yearbook 1973. 0.8. Dept. of Health, Education and Welfare/F.A.O. 1972. Food Composition table for use in East Asia. Van Buren, J. P.; L. R. Hackler; and K. H. Steinkraus. 1972. Solubilization of soybean Tempeh consti- tuents during fermentation. Cereal Chemistry 49:208. Van Damme, P. A.; A. G. Johannes; H. C. Cox; and W. Berends. 1960. On Toxoflavin, the Yellow Poison of Pseudomonas cocovenenans. Rec. Trav. Chim. 79:255. Van Veen, A. G., and W. K. Mertens. 1934. Toxic consti- tuents of the so-called Bongkrek Poisoning in Java. Rec. Trav. Chim. 53:257-266. Chemical Abstr. 28, 1934. Van Veen, A. G. 1935. Bongkrekic Acid. A blood-sugar lowering substance. Rec. Trav. Chim. 54:373—380. Chemical Abstr. 29, 1935. Van Veen, A. G. 1936. Influence of Bongkrekic acid on carbohydrate metabolism. Arch. Neerland. Physiol. 21:73-103. Chemical Abstr. 30, 1936. Van Veen, A. G., and G. Schaefer. 1950. The influence of Tempeh fungus on the soya bean. Documenta Neerl. et Indonesica de Morbis Tropicis 2:270. Wagenknecht, A. C.; L. R. Mattick; L. M. Lewin; D. B. Hand; and K. H. Steinkraus. 1961. Changes in soybean lipids during Tempeh fermentation. J. Food Sci. 26:373. Wang, H. L.; D. I. Ruttle; and C. W. Hesseltine. 1969. Antibacterial compound from a soybean product fer- mented by Rhizopus oligosporus (33930). Proceedings of the Society for Experimental Biology and Medicine 31:579. Wang, H. L.; E. W. Swain; and C. W. Hesseltine. 1975. Mass production of Rhizopus oligosporus spores and their application in Tempeh fermentation. J. Food Sci. 40:168. Wheeler, E. L., and R. E. Ferrel. 1971. A method for phytic acid determination in wheat and wheat fractions. Cereal Chemistry 48:312. Wolf, W. J. 1970. Scanning electron microsc0py of soybean protein bodies. J. Am. Oil Chem. Soc. 47:107. APPENDICES APPENDIX A FATTY ACIDS Carbon atoms and double bonds APPENDIX A FATTY ACIDS Systematic name Butanoic Hexanoic Octanoic Decanoic 4 Decenoic 9 Decenoic Dodecanoic 4 Dodecenoic 9 Dodecenoic Tetradecanoic 4 Tetradecenoic 5 Tetradecenoic 9 Tetradecenoic Hexadecanoic 9 Hexadecenoic Octadecanoic 6 Octadecenoic 138 Common name Butyric Caproic Caprylic Capric Obtusilic Caproleic Lauric Linderic Lauroleic Myristic Tsuzuic Physteric Myristoleic Palmitic Palmitoleic Stearic Petroselinic Carbon atoms and double bonds 139 Systematic name C18:1 C18:1 C18:2 C18:3 C18:3 C24:0 C24:1 c26:1 30:1 9 Octadecenoic 11 trans—Octadecenoic cis-cis—9,12, Octadecadienoic cis-cis-cis-9,12,15 Octadecatrienoic cis—trans-trans- 9,11,13, Octadecatrienoic trans-trans-trans 9,11,13, Octadecatrienoic 9,11,13,15, Octadecatetraenoic Eicosanoic 9, Eicosenoic 11, Eicosenoic 5,8,11,14 Eicostatetraenoic Docosanoic ll Docosenoic l3 Docosenoic 4,8,12,15,19 Docosapentaenoic Tetracosanoic 15 Tetracosenoic 17 Hexacosenoic 21 Triacontenoic Common name Oleic Vaccenic Linoleic Linolenic Eleostearic Eleostearic Parinaric Arachidic Gadoleic Arachidonic Behenic Cetoleic Eurcic Clupanodonic Lignoceric Selacholeic Ximenic Lumequeic APPENDIX B TEMPERATURE OF TEMPE DURING FERMENTATION AT 25°C AND 32°C APPENDIX B TEMPERATURE OF TEMPE DURING FERMENTATION AT 25°C AND 32°C Time (hrs) At 25°C At 32°C 0 25.0 25.0 2—3/4 25.0 30.0 7 25.0 30.5 14 25.0 31.5 17 25.0 33.0 18 25.0 33.5 19 25.0 35.0 19.5 25.0 35.5 19.75 25.0 36.0 20 25.0 36.5 20.25 25.0 37.0 20.5 25.0 37.5 20.75 25.0 38.0 21 25.0 38.0 21.5 25.0 39.0 22 25.0 40.0 22.5 25.0 40.0 22.75 25.0 40.5 23 25.0 41.0 24 25.0 42.0 25 25.0 42.5 26 25.0 43.0 26.5 25.0 43.5 27 25.0 43.5 140 fl. .. 141 Time (hrs) At 25°C At 32°C 27.5 25.0 42.5 28 25.5 42.5 28.5 25.5 42.5 39.5 28.0 40.0 40.0 28.5 40.0 41.0 29.0 39.5 APPENDIX C SENSORY RANCIDITY TEST APPENDIX C SENSORY RANCIDITY TEST Please indicate the rancidity of the following numbered soybean samples, by smelling an open jar, one at a time. If the sample is rancid, please indicate the magnitude of its rancidity. Sample No. RanCidity None Weak Strong 142 APPENDIX D ANIMAL FEEDING LOG a.. .o. n.. .o- .u. ..o o.. o.. .n. cu. o.. .c. .o. .n. on. coo no. no. .co oo. ..o o.. on. o.. ........anww.m to. at. can no. no. no. OD. CC. 0.. DO. 00- .0. .00 non no. so. no. on. on. once no. 000 on. at. ........"Umwh Show wumc mumw cognaoo comm comm cwummz Hwawmucoo Mozamucoo .0: H952 .ummno .uflcH + + um u< owmm ucwmm comm HanuacH Damme: Hasncm .............. TUMU COHum>waAO moq OGHUTOW Hmfifldd ......... wuwfl .HMCH 00a DZHDmmm A<2H2< Q XHQmem4 143 APPENDIX E STANDARD CURVE FOR Fe Ha ooa\ms ca mm mo :Owumuusoocoo mN.o om.o _ 1 Oh mom m>mDU QK