. I 5.4%.an ‘ 9v. , kw. :5: a r imwumwvmia 1. may ......u.. ‘ Q I h gnasfilsl ”$90."... “up! .. r1}. 2Lhu‘r Suntfinfinhm :sl; :4. \Y. t:- v ! 1‘31 .337 .. t ..I.V.PL..3IS.J. 1.7 2:. 2.1:. at. P. . .s t! , ‘v‘EZ I U:- .32. l’i' .‘u '1 , .t .. .33.... 5.73.. 5.4% .r... ‘ . . . .. .. a u. x 4.5. 3 . «22%. L :0 3.3”»). :5 V . , , . . . P .3... N £32.. .. . y. i. 1:. 1 , .C'til:t. , ‘ . i , , uh . m n. . 5 5.3;. y L! l '1: 1" 9.13.1.3 33.4....“ .. (I. .. .1 v, . ... 191....1? dif- .335: ‘ .2 I. . {6.1. Ht... u 5.». , . , av ‘53:}. 1... : : . , an“ huh... .8 .2. .2... 1 . . ." V . V . :3. 3. 2:; , , V ‘ 25 . EB . .2 no! . 3.10.: 1.. .. I) q»: . i205~vau va... A: , 2..)\.. h 1.1. .fl .: q 5...: 3i .......fii..ts....! . E 1M. “3 .ll. . V .3. 1 4.35; y. .F. x 2 Sr 3.8:. at} :- Aim... Hugh» .LIHVN... I - I... s ‘ 0.. so I. g .. .1-) .12 . u. .. .2. 2.- :21: 5.val. Iiia «szix {it'llsW‘O (.10.) t 3:9.»2b1... . dark“. at-» ,. "433 5...!) . ..., é , 1.3.33“ l??? 43...! “v 1 :3. .r. . “1.5.IIL. .. 3.0 4 t 2...: 37.25.! c! .2 : . , , , ‘ .71 3Q.I.Al(‘!\(..¥l 3.1. llllllllllfllllWmnwwlni LIBRARY Mlchlgan State Unlverslty This is to certify that the thesis entitled Extrusion Processing of Cereals and Legumes: Protein-Phytate-Iron Interactions in Relation to Iron Bioavailability presented by Padmashri Ummadi “"“ has been accepted towards fulfillment of the requirements for Ph .0. degree in Human Nutrition WW Major professor DateW 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE ll RETURN BOXtomavombchockouMyourncud. ‘ TO AVO|D FINES Mum onorbdonddoduo. DATE DUE DATE DUE DATE DUE MAGC' . ' I Mfim‘ilrzooj- MSU I. An Affirmative MM“ Opponmlly twat Ell ___w fi‘ EXTRUSION PROCESSING OF CEREALS AND LEGUNES: PROTEIN-PEYTATE-IRON INTERACTIONS IN RELATION TO IRON EIOAVAILAEILITY BY Padmashri Ulnadi 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 1994 ABSTRACT EXTRUSION PROCESSING OF CEREALS AND LEGUMES: PROTEIN-PHYTATE-IRON INTERACTIONS IN RELATION TO IRON BIOAVAILABILITY BY Padmashri Ummadi The effects of extrusion processing on availability of iron from cereals and legumes was investigated. Legume flours extruded at 110°C had higher in vitro iron dialyzability compared to those extruded at 135°C or to boiled legume flours. Durum wheat semolina extruded at 50°C or 96°C had higher amounts of dialyzable iron than raw semolina. In vivo iron bioavailability, measured by rat hemoglobin regeneration efficiency, was lower in durum wheat pasta extruded at 205°C compared to semolina or a standard source of iron, ferrous sulfate. The negative effect of high temperature extrusion on iron availability was diminished by navy bean supplementation. Durum.wheat (85%)/navy bean flour (15%) pasta extruded at the same temperature (205°C) had higher iron bioavailability than durum wheat pasta. The mechanisms involved in the effects of extrusion processing' on iron. availability' were studied» .AIthough extrusion resulted in extensive degradation of phytic acid and tannins, two known inhibitors of iron bioavailability, these factors alone could not explain the varying extrusion effects on iron dialyzability. A more complex system involving the associations of protein, phytate, and iron was investigated by in vitro assays. Sodium dodecyl sulfate-polyacrylamide gel electrophoretic patterns of reduced and unreduced durum wheat protein fractions indicate that the formation of disulfide linkages during extrusion may be responsible for the greater thani two-fold increase in insoluble jprotein fraction of semolina after extrusion at 50°C or 96°C. Extrusion at both experimental temperatures also increased the amount of iron and phytate associated with the albumin and globulin fractions of semolina. An iron binding polypeptide (MW:16,500 Da) was identified in the albumin, globulin and insoluble protein fractions of durum wheat. It is proposed that an increase in the iron bound to the soluble and low molecular weight proteins, albumins and globulins may explain the increased dialyzability of iron after low temperature extrusion. ACKNOWLEDGMENTS I gratefully acknowledge the financial support of the John Harvey Kellogg Fellowship which enabled me to complete my doctoral program. I would like to express my gratitude to my major professor, Dr. Wanda Chenoweth for her thoughtful guidance, encouragement and advice throughout my graduate study. I also would like to thank my committee members, Dr. Maurice Bennink, Dr. Perry'Ng, Dru Mark‘Uebersax and Dr. Irvin Widders for serving on my committee and offering their technical expertise. Thanks to my labmates and all my friends at Michigan State University for making my stay here exciting and memorable. My love and gratitude to my parents and family members for inspiring and stimulating my interest in learning. Finally, my deepest appreciation is extended to a special person in my life, Rishi Nalubola for his confidence in my abilities and his unflinching support towards my graduate education. iv TABLE OF CONTENTS Page LISTOFTABLES0.0000000....00....000.000.000.0000000000 Vii LISTOFFIGURES000......0.00.00.00.00... ......... 0 ..... ix CHAPTER 10 INTRODUCTION 00.0.0000... 00000000 .0 ....... .00 1 Literature Review ........................... 4 Extrusion processing .................... 4 Chemical interactions of minerals inf00d80000000 00000 000.000.00.000. 5 Characterization of iron in cereals andlegumes0.000.000.00000000000000 7 Determining iron bioavailability ........ 9 Effect of protein on iron and zinc bioavailability .................... 10 Effect of phytate and phytate-protein complexes on iron and zinc bioavailability .................... 19 Effect of extrusion processing on mineral bioavailability ............ 24 Effect of extrusion processing on physiochemical properties of proteins ........................... 26 Wheat.proteins .......................... 29 Justification ............................... 33 References .................................. 36 CHAPTER 2. THE INFLUENCE OF EXTRUSION PROCESSING ON IRON DIALYZABILITY, PHYTATES AND TANNINS INLEGWES0.00.000.00.000.00000000000.0.0.0. 47 Abstract ............... ....... .............. 48 Introduction ................................ 49 Materials and.Methods ....................... 51 Results and Discussion ...................... 56 References ........... ......... ..... ......... 66 CHAPTER 3. CHAPTER 4. CHAPTER 5. CHAPTER 6. CHAPTER 7. BIOAVAILABILITY OF IRON IN EXTRUDED WHEAT PRODUCTS 000.00.000.0000000.0.0....0000. 00000 Abstract ................ ....... ............. Introduction ............ .................... Materials and.Methods ....................... Results and Discussion ...................... References .................................. PROTEIN-PHYTATE-MINERAL INTERACTIONS IN EXTRUDEDPRODUCTS0000.0000000000000000 000000 Introduction .......... . ..................... Materials and Methods .... ........ ........... Results and Discussion ...................... References .................................. EXTRUSION PROCESSING OF SEMOLINA. I. CHANGES IN THE SOLUBILITY AND DISTRIBUTION OF PROTEINS 00000.0000.000000000000000...00 00000 Abstract ............. .............. ......... Introduction ................................ Materials and.Methods ....................... Results and.Discussion ...................... References .................................. EXTRUSION PROCESSING OF SEMOLINA. II. DISTRIBUTION OF IRON AND PHYTATE IN PROTEIN FRACTIONS00.0..0.000.000000.0000000000000000 Abstract .................................... Introduction ................................ Materials and.Methods ....................... Results and Discussion ...................... References .............. ................... . SUMYANDCONCLUSIONS 0.0000000000000000... Recommendations for Future Research ......... References .0.0.0.000000000000000000000.0.... vi 68 69 70 71 77 85 87 88 89 93 100 101 102 103 104 106 116 117 118 119 122 125 135 137 142 143 LIST OF TABLES Table CHAPTER 2 1. Total iron and dialyzable iron in boiled and extruded legume products .......................... 2. Nonelemental, soluble and ionic iron in boiled and extruded legume products ...................... 3. Degradation products of phytic acid in raw, boiled, and extruded legume products .............. 4. Tannin content (mg catechin equivalents/g dry weight) of raw, boiled, and extruded legume products ...... CHAPTER 3 1. Compositions of diets (g/lOOg) used in animal study ............................................. 2. Total iron and dialyzable iron (D-Fe) content (pg/g dry weight) of durum wheat flour, semolina, wheat pasta, wheat (85%)/navy bean (15%) pasta and navy bean flour ........................................ 3. Iron bioavailability of extruded and non-extruded products in rats using the hemoglobin repletion technique ......................................... 4. Statistical analyses of data obtained from rat study ...................................... ....... CHAPTER 4 1. Changes in solubilities of protein, phytate, iron and zinc in durum wheat flour caused by extrusion, dephytinization and dialysis with pepsin and pancreatin ........................................ 2. Changes in solubilities of protein, phytate, iron and zinc in wheat bean pasta and navy bean flour caused by dephytinization and dialysis with pepsin and.pancreatin ....... ..... ........................ vii 57 59 61 63 74 78 80 81 94 98 CHAPTER 5 1. Protein fractionation of raw and extruded semOIina0..00....0.0...00000...0.00.00.00.00...000 107 CHAPTER 6 1. Iron in protein fractions from raw and extruded semolina .......................................... 126 2. Phytate in protein fractions from raw and extruded semolina .......................................... 129 viii LIST OF FIGURES Figure Page CHAPTER 4 1. Flow chart for analyses of soluble protein, phytate, iron and zinc in raw and extruded products ........ 91 CHAPTER 5 1. Sodium dodecyl sulfate polyacrylamide gel electro- phoretic patterns of unreduced and reduced (with 2-mercaptoethanol) albumin, globulin and gliadin fractions from raw and extruded semolina; Np-protein extract from flour of cultivar Neepawa; 2,5,8,11,14,17 represent fractions from raw semolina; 3,6,9,12,15,18 represent fractions from semolina extruded at 50°C; 4,7,10,13,16,19 represent fractions from semolina extruded at 96°C; HMW-GS - high molecular weight glutenin subunits .......................................... 109 2. Sodium dodecyl sulfate polyacrylamide gel electro- phoretic patterns of unreduced and reduced (with 2-mercaptoethanol) glutenin and insoluble fractions from raw and extruded semolina; Np-protein extract from flour of cultivar Neepawa; 2,5,8,11 represent fractions from raw semolina; 3,6,9,12 represent fractions from semolina extruded at 50°C; 4,7,10, 13 represent fractions from semolina extruded at 96°C; HMW-GS - high molecular weight glutenin subunits ................................... ....... 112 CHAPTER 6 1. Iron staining of reduced (with 2-mercaptoethanol) albumin, globulin and insoluble fractions from raw and extruded semolina; Hb-human hemoglobin (reduced with 2-mercaptoethanol); 2,5,8 represent fractions from raw semolina; 3,6,9 represent fractions from semolina extruded at 50°C; 4,7,10 represent fractions from semolina extruded at 96°C .......... 132 ix CHAPTER 1. INTRODUCTION 2 Iron is a trace element that is of continuing public health concern. Iron deficiency is a common disorder with incidence varying widely with age, sex, race and economic status (Fairbanks et a1., 1971; DeMaeyer and Tegman, 1985; Stoskman, 1987; Arthur and Isbister, 1987; Skikne, 1988). The prevalence of anemia is estimated at 30% of the world population (DeMaeyer and Tegman, 1985). Nutritionists have routinely expressed concern that the levels of iron intake are not adequate and marginally deficient intakes may exist (Wolf, 1982; Arthur and Isbister, 1987). Food iron may be classified as either heme iron or nonheme iron. Although the percent absorption of nonheme iron is much lower than that of heme iron, a normal diet is higher in nonheme iron than in heme iron. Thus, the major contribution of available iron is made by nonheme iron from sources such as grains, legumes, fruits and vegetables. Heme iron sources include animal foods such as beef, pork, lamb, chicken, fish and liver. A meal is a complex, chemical mixture containing compounds which interact with minerals and either inhibit or facilitate their absorption. Total mineral content in a food or meal, therefore, provides only a rough approximation of the amount available for absorption (Monsen, 1980). Bioavailability is defined as the proportion of the total mineral in a food, meal or diet that is utilized for normal body functions (Fairweather-Tait, 1992). Variations in 3 dietary levels and forms of proteins, peptides, and amino acids can affect the bioavailability of a variety of minerals including iron and zinc. In addition, during food processing, interactions that take place between minerals and other components of food such as proteins and phytates may have either a positive or negative effect on mineral bioavailability. LITERATURE REVIEW EXTRUSION PROCESSING The food extruder is considered a high-temperature short- time bioreactor (Harper, 1989) that.is being used increasingly to manufacture a variety of products. These include increasing varieties of ready-to-eat cereals, salty and sweet snacks, croutons for soups and salads, dry and soft-moist pet foods, pasta and macaroni products, texturized meat analogs made from vegetable protein, precooked food mixtures for infant feeding, soup and drink bases, and pregelatinized starch. The operation of an extruder has been summarized by Harper (1989). Ingredients are released from the feed hopper into the preconditioner at a controlled rate. The preconditioner is a pressurized chamber in which raw granular food ingredients are uniformly moistened or heated by contact with water or steam before entering the extruder. When the food enters the extruder, the extrusion screw sequentially conveys and heats food ingredients and works them into a continuous plasticized mass while rotating in a tightly fitting barrel. As the flights on the extruder screw convey the food materials down the barrel, the mechanical energy used to turn the screw is dissipated causing a rapid rise‘in the temperature of the food ingredients. The resulting plasticized feed ingredients are then forced through a die. 5 The pressure drop across the die rapidly converts the high- temperature water in the product to steam and causes puffing to occur. The use of twin-screw extruders is relatively new and has the advantages of improved conveying and mixing capabilities and an extended range of applications compared to the single- screw extruder (Smith and Ben-Gera, 1980). The advantages of extrusion processing over conventional cooking procedures are its relatively low cost, high productivity, energy efficiency, ability to produce high quality products, ability to process dry, viscous materials, and ease of production of new foods and product shapes (Harper, 1989). CHEMICAL INTERACTIONS OP MINERALS IN FOODS There are many reactions that minerals might undergo in foods to effect a change in solubility and/or charge that subsequently alters their bioavailability. Metals exist in ‘water not as ions but.as hydrates. IHydrolysis of the hydrates may occur as the pH is raised, causing them to lose protons and forming less soluble or insoluble hydroxides, which may precipitate and thus become unavailable (Clydesdale, 1988). Various types of bonds can effectively tie up and precipitate or solubilize a mineral in a complex. Chelation is one form of complex formation where the ligand forms more than one bond with thetmineral. Mineral complexes can also be formed through ionic and covalent bonds. Often, these bonds 6 are responsible for the chemical properties of the complex. In determining the physical properties, including solubility of a complex, intermolecular forces such as dipoles, hydrogen bonds, and dispersion or London forces play a more important role than bonding (Clydesdale, 1988). The mechanisms involved in mineral interactions in foods which have the potential to affect bioavailability have been broadly defined into four categories by Clydesdale (1989): 1. Mineral Displacement - Displacement of a mineral from a complex with another mineral to form a soluble (available) or insoluble (unavailable) complex, eg. mineral reactions with fibers or phytate. 2. Polymineral-Ligand Complexes - The addition of a second or third mineral to a soluble mineral-ligand complex causing precipitation by forming a polymineral-ligand complex, eg. calcium-zinc-phytate complex. 3. Polymineral-Polyligand Complexes - The addition of a mineral causing a mineral-ligand complex to form one or more minerals binding to more than one substrate (ligand) and forming* a polymineral-polyligand complex, eg. iron-zinc- phytate-protein complex, protein-phytic acid-zinc complex. 4. Enzyme Susceptibility of Complexes - The formation of a polymineral-ligand complex which changes the susceptibility of the mineral-ligand bonds to cleavage by digestive enzymes. 7 CHARACTERIZATION OF IRON IN CEREALS AND LEGUNES The form in which iron exists in wheat has been studied. In wheat bran, approximately 60% of the total iron is present in a salt extractable form identified as monoferric phytate. The biological availability of iron from monoferric phytate, either isolated from wheat bran or the synthetic product, determined by a hemoglobin depletion-repletion bioassay, was seen to be equal to a reference compound, ferrous ammonium sulfate (Morris and Ellis, 1976). Another form of iron, ferric phytate (3 to 4 moles iron/mole phytate) , thought to be present in the insoluble bran residue, had significantly lower biological availability than monoferric phytate or the reference compound. Because of its salt-extractability and water soluble properties, it was proposed that monoferric phytate in bran may be bound to cationic sites of proteins or other cellular components and the utilization of the iron is through solubilization of the monoferric phytate by an ion exchange mechanism (Morris and Ellis, 1976). In another study, May et al. (1980) examined the nature of the endogenous iron in wheat grains and bran using 57Fe Mbssbauer spectroscopy. The spectra of seeds and bran from wheat grown in 57Fe-enriched culture indicated that most of the iron in wheat.is present.in the form of monoferric phytate with a high-spin configuration. Iron in different legumes has also been isolated and characterized. Crichton et al. (1978) isolated phytoferritin, 8 a storage form of iron, from pea and lentil seeds. Subunits with molecular weights of 20,300 and 21,400 Da were identified in peas and lentils, respectively. Phytoferritins were seen to have a larger cavity in the interior of the molecule than mammalian ferritin, thus enabling them to store 1.2 to 1.4 times as much iron. Although the quartenary structure of the phytoferritins was similar to mammalian ferritin, there were large differences in primary structure, phytoferritins having a higher asparagine/aspartic acid content and a higher isoelectric point than mammalian ferritin. Phytoferritin was identified by Lynch and Covell (1987) as the major iron- containing protein fraction in commercial soybean flour. Kojima et al. (1981) studied iron distribution in pinto beans in terms of ease of solubilization. They reported that about 25% of the iron in pinto beans is readily soluble upon incubation, 45% of the iron can be mobilized by chelating or reducing agents and 30% of the iron is firmly bound to the insoluble bean residue. Chidambaram et al. (1987) reported fractionation and partial characterization of iron binding components of digested pinto beans. In vitro enzymatic digestion and dialysis (molecular weight cutoff of 14,500 Da) showed that 58% of the iron binding components of pinto beans are dialyzable. Using 59Fe, a low MW (14,500 Da) protein fraction was identified in the dialysate. The predominant non- dialyzable protein fraction, which retains 42% of iron was 9 found to have a MW of 49,000 Da. One of the weaknesses of this study is the use of 59Fe to identify iron binding proteins which requires the assumption of equilibration between the added radioactive iron and the naturally present iron. The major hindrance to studies identifying or quantifying iron binding protein fractions from foods is the lack of appropriate techniques. One alternative to using radio-iron is the use of reagents that can specifically stain iron binding proteins separated on a polyacrylamide gel. Chung (1985) developed a simple and fast staining procedure using a Ferene S/thioglycolic (acid. reagent 'that can. be used. to specifically stain iron binding proteins on polyacrylamide gels. The method is based on the reaction between Ferene S and iron atoms present in the proteins with thioglycolic acid acting as a reducing agent (converting Fe+++ to Fe”) and an anion labilizer (facilitates iron release from the proteins). The reagent was shown to detect transferrin, lactoferrin, ferritin, hemoglobin and cytochrome c with good sensitivity. DETERHINING IRON BIOAVAILABILITY Several approaches including in vitro, animal and clinical studies have been used to estimate the bioavailability of iron in foods. One in vitro method is based on the assumption that a soluble form of iron is likely to be absorbed. It includes the measurement of iron soluble 10 in dilute hydrochloric acid (Shah et al., 1977; Lee and Clydesdale, 1979). The most common in vitro procedure is the iron dialyzability assay (Miller et a1., 1981). Dialyzable iron passes from an in vitro enzymatic (pepsin-pancreatin) digestion of a food or test meal into the interior of a dialysis bag containing a bicarbonate buffer. The most widely used animal method compares the hemoglobin response in anemic rats given graded quantities of the iron source with that obtained from comparable amounts of ferrous sulfate. Two ‘variations of the:method.are possible -- the slope ratio assay (official method of AOAC) which measures the relative biological value of an iron source, and the hemoglobin regeneration efficiency (Mahoney and Hendricks, 1982) which measures the percentage of iron ingested that is incorporated into hemoglobin. Clinical studies in humans to determine the availability of iron in foods commonly involves feeding of a food with extrinsic or intrinsic radioiron tags. At the end of the experimental period, blood is drawn to measure hematocrit and serum ferritin, and to determine the absorption of iron (Cook et al., 1972; Hallberg and Bjorn-Rasmussen, 1972; Forbes et al., 1989). EFFECT OF PROTEIN ON IRON AND ZINC BIOAVAILABILITY Protein-Iron Interactions A great deal of literature is available on the interactions between proteins and iron during absorption and 11 their effects on the availability of the mineral. A proposed mechanism for nonheme iron absorption suggests that free iron released during digestion chelates with a variety of substances, including proteins. Some of these substances bind iron tightly and are excreted along with the bound iron, while other substances release the iron at the mucosal surface (Monsen, 1988). Results from studies using extrinsic radioactive iron labels added to single foods or to meals have shown that iron absorption differs depending on the type of protein in food. Noncellular proteins such as egg albumen, milk, cheese, and soy protein tend to depress nonheme iron absorption in humans (Cook and Monsen, 1976; Lynch et al., 1985) while cellular proteins such as beef, liver, lamb, pork, chicken, fish are known to enhance nonheme iron absorption in humans (Hallberg et al., 1979; Cook and Monsen, 1976). In addition to their effects on non-heme iron, cellular proteins also have been shown to enhance heme iron absorption in humans (Hallberg et al., 1979). The Meat Factor. The facilitation of nonheme iron absorption by meat is not a general property of all animal tissues but is specific to certain animal proteins (Hurrell et al., 1988). The identity of this iron absorption enhancing "factor" in meat is not known in spite of considerable research. It is suggested that the meat factor may be the cysteine-containing peptides released during digestion of beef 12 (Martinez-Torres et al. , 1981; Taylor et al. , 1986) . Cysteine containing peptides produced during meat digestion were seen to enhance absorption of nonheme iron in humans. In contrast, oxidizing these products reduced the absorption of nonheme iron by 63% (Taylor et al., 1986). Thus it appears that free sulfhydryl groups aid in the absorption of nonheme iron. Meat may enhance iron bioavailability by keeping iron in a soluble form. The iron solubilization effect of meat may be due to a particular peptide fragment(s) and/or profile of amino acids. In vitro pepsin digestion products with MW <10,000 solubilized significantly more iron than those with MW >10,000 (Clydesdale, 1988). Both raw and cooked meat appear to increase nonheme iron absorption. A water extract of cooked beef’ has been :reported. to increase nonheme iron absorption. to a lesser' extent. than. the residue (Bjorn- Rasmussen and Hallberg, 1979). Monsen (1988) reported that addition of a mixture of amino acids that mimic beef increased the absorption of nonheme iron. The Milk Factor. Based on in vitro studies conducted using a mixture of milk with iron in cereals (Clydesdale and Nadeau, 1984) and with iron in model systems (Platt et al., 1987) , Clydesdale ( 1988) proposed that there is a "milk factor" which exerts a protective effect against the fiber- and phytate-induced precipitation of iron and thus increases its potential bioavailability. Clydesdale and Nadeau (1984) evaluated the ability of milk and several of its fractions to 13 solubilize iron in corn, wheat and oats. There was more total soluble iron in the presence of whole milk, lactose-free milk and nonfat dry milk than a water control. When deproteinized milk was used, the solubilization effect largely disappeared in wheat and in a mixture of the three-grains. They concluded that the milk factor seemed to be in the protein fraction. In vivo studies have also demonstrated the iron absorption enhancing effect of milk. Randhawa and Kawatra (1993) found that a habitually consumed diet supplemented with 190 ml milk (equivalent of 8 g protein) per day greatly improved the apparent absorption and retention of iron, zinc, copper and manganese in pre-adolescent girls. Egg Proteins. Peters et al. (1971) reported low bioavailability of egg yolk iron in humans as well as in vitro. Phosvitin, the principal phosphoprotein of the yolk, contains about 0.4% iron, accounting for most of the iron present (Taborsky, 1983). It has been shown that 50% or more of the amino acid residues of this protein are phosphoserine, accounting for the high iron-binding capacity (Taborsky, 1983). Sato et al. (1984) showed that the phosvitin-iron complex could promote iron precipitation in the intestine. Albright et al. (1984) reported that iron from the phosvitin- iron complex is not released by heating or treating with citric acid or NaCl. Only EDTA released iron from phosvitin without heating. Phosphopeptides released from phosvitin during luminal digestion of egg yolk protein have the ability 14 to strongly bind ferric iron in the lumen (Sato et al. , 1985) . Thus, phosvitin is presumed to be responsible for the poor availability of egg yolk iron. Soy Proteins, Legume proteins are generally acknowledged to adversely affect iron bioavailability (Bothwell et al., 1982) . Soybeans are rich in iron, but the bioavailability of iron in humans is low. One explanation may be that a major fraction of the iron in soybeans is present in the form of phytoferritin which is known to have low bioavailability (Lynch and Covell, 1987). Substitution of a portion (30% to 50%) of meat in a hamburger meal by soy flour reduced nonheme iron bioavailability in humans (Hallberg and Rossander, 1982; Lynch et al., 1985) . In vitro studies suggest that the inhibitory effect of soybean products on nonheme iron bioavailability may be related to the protein component of soy products (Schricker et al., 1982). In vitro digestion of soybeans showed iron trapped within large peptide aggregates; iron was released only upon dissociation of the aggregates (Schnepf and Satterlee, 1985) . A soybean protein isolate reduced nonheme iron absorption to a greater extent than the whole beans and extensive enzymic hydrolysis of the soybean protein with papain removed most of the inhibitory effect (Lynch and Covell, 1987). The influence of soy protein on iron nutrition, however, is not entirely negative. When soy flour was partially substituted for beef, there was 30-60% improvement in heme 15 iron absorption compared to a beef control. This increase in the absorption of heme iron was seen when soy flour was incorporated with ground beef into a meat patty and fed to human subjects solely as a patty or as part of a mixed meal (Lynch et al., 1985). Not only the iron naturally present in foods, but also the iron added to foods can be affected by the type of protein in the food. It has been shown that bioavailability of added iron can be influenced by the formation of iron complexes with protein digestion products. In a study measuring iron dialyzability in the presence of selected proteins and fractionated protein digestion products, Kane and Miller (1984) found that dialyzability of added iron (ferric chloride) was higher for bovine serum albumin and beef, than for egg albumin, gelatin, casein, soy protein isolate or gluten. Low molecular weight (<6000 to 8000 Da) digestion product fractions from bovine serum albumin and beef were seen to enhance iron dialysis; similar fractions from casein and soy protein isolate did not affect iron dialysis. The authors speculated that the influence of proteins on iron bioavailability may be related to the affinity of undigested or partially digested protein for iron and to the stability of small molecular weight soluble iron complexes formed from protein digestion products. Another study conducted by Nelson and Potter (1979) determined, in vitro, the ability of five protein sources 16 (wheat gluten, soy protein isolate, zein, albumin, and casein) to bind added ferrous and ferric iron as well as the effects of pH, time and temperature on this binding; More than 50% of the added iron (ferrous or ferric iron) was bound to the insoluble fraction of the proteins. In vitro digestion of these complexes in an HCl-pepsin-pancreatin system released 64-97% of the iron suggesting that protein-bound iron may be readily freed for absorption within the gastrointestinal tract. Subsequent research using the hemoglobin repletion assay in rats showed that protein-bound ferrous iron was as biologically available as ferrous sulfate. In contrast, the bioavailability of protein-ferric complexes was lower than the standard, ferric pyrophosphate (Nelson and Potter, 1980). Protein-zinc Interactions Changes in dietary protein levels affect the absorption of a number of minerals including zinc. However, the effect of dietary protein on zinc absorption in humans is controversial. Greger and Snedeker (1980) observed that human subjects absorbed zinc more efficiently when fed a high protein (24.1g N/day) diet than a low protein (8.1g N/day) diet. Other investigators have observed either improvements in zinc absorption and/or retention (Price et al., 1970) or no changes (Colin et al., 1983) when dietary protein levels were elevated. 17 Several individual amino acids, particularly histidine, cysteine, and tryptophan, have been found to increase absorption and/or retention of zinc in tissues of rats (Greger and Mulvaney, 1985; Wapnir and Stiel, 1986). In humans the bioavailability of zinc from zinc:histidine complexes mixed in a ratio of 1:2 or 1:12 was shown to be higher when compared to zinc sulfate (Scholmerich et al., 1987). Theories on how amino acids may improve zinc absorption have been put forth by many investigators. Greger (1988) suggested that amino acids may improve absorption by forming soluble complexes with zinc and thus prevent formation of insoluble hydroxides. Wapnir and Stiel (1986) studied, in vivo, the structural characteristics of amino acids that could influence the facilitation of zinc intestinal absorption. Greater zinc absorption from zinc-amino acid complexes than from zinc-non amino acid homologues in the small intestine of rats suggests that amino acid transport systems may be involved. Dietary proteins can also affect mineral utilization when minerals are "trapped" inside protein or peptide complexes that are resistant to proteolysis. Lonnerdal (1987) suggested that incomplete hydrolysis of casein in cow's milk by infants may lead to decreased absorption of zinc. Browning reaction products (amino acid-carbohydrate complexes) in toasted and other heat-treated foods which are resistant to hydrolysis in the gut can complex metals and have been shown to reduce zinc absorption (Lykken et al., 1986). Similarly, it has been 18 hypothesized that when protein-phytate-zinc complexes are incompletely digested, mineral absorption is decreased (Erdman et al., 1980). Although a great deal of research has focused on the effects of specific proteins on iron utilization, the effects of egg and meat proteins on zinc availability have not been studied in depth” The effect of milk protein on the absorption of zinc from sources other than milk has not been studied. However, much research has been done on the absorption of the zinc present in milk itself. Studies comparing human milk and cow milk have consistently shown that zinc is absorbed more efficiently from human milk (Sandstrom et al., 1983). Bobilya et al. (1991) found that in neonatal pigs the bioavailability of zinc in non fat dry milk and low fat plain yogurt was very high and similar to that in zinc carbonate. Wood and Hanssen (1988) reported that compared to water, cow's milk and lactose-free milk significantly reduced zinc absorption in postmenopausal women. Pecoud et al. (1975) showed that when single.doses of zinc sulfate were given along with milk to healthy young volunteers, zinc absorption was decreased as indicated by a significant drop in serum zinc levels. Zinc is known to form complexes with soy proteins-also. In a study of zinc complexes, the major portion of extractable zinc from defatted soy flour was found to be either free or associated with very low molecular weight proteins, peptides 19 or their complexes with phytic acid rather than the major proteins of soybean (Clydesdale, 1989) . The interactions between zinc, proteins and phytic acid are discussed further in the section below dealing with phytate-protein-zinc complexes. EFFECT OF PHYTATE AND PHYTATE-PROTEIN COMPLEXES ON IRON AND ZINC BIOAVAILABILITY Phytate [myoinositol 1,2,3,5/4,6-hexakis (dihydrogen phosphate)] is a naturally occurring organic compound found in plants. It constitutes 1-6% by weight of most nutritionally important legume, cereal, and oilseeds, where it serves as a storage phosphate that becomes available during germination (Lasztity and Lasztity, 1990). Phytic acid strongly interacts with proteins in a pH- dependent manner. At low pH, phytic acid forms electrostatic linkages with the basic lysine, arginine, and histidine residues, resulting in neutral, insoluble complexes that dissolve only below pH 3. At neutral and basic pH, both phytate and most proteins have a net negative charge which leads to their virtually complete dissociation from each other. In the presence of multivalent cations, however, protein-cation-phytate complexes may occur. These complexes tend to dissociate at extremely high pH (>10) and the phytic acid becomes insoluble, while the protein remains in solution (Cheryan, 1980). 20 O'Dell and deBoland (1976) developed a method for the identification of protein-phytate complexes on polyacrylamide gels via the precipitation of phytate as a white band of ferric phytate. Recently, DiLollo et al. (1991) showed that the method developed by O'Dell and deBoland is not specific for protein-phytate complexes. They evaluated the use of a chromogen, cobalt hexaammine chloride solution believed to be specific for iron, as a means to improve the visualization of the ferric phytate on the gels. Results suggest that the white bands which are believed to indicate protein-phytate complexes could be the result of non-specific protein iron interaction in addition to phytate iron interaction. These findings were supported by experiments in which proteins containing no phytate gave positive staining for the reaction believed to be specific for a phytate-iron complex. A procedure to accurately identify protein-phytate or protein- phytate-iron complexes separated on polyacrylamide gels is yet to be developed. Phytate has been suggested to influence protein digestibility. Ritter et al. (1987) reported that in vitro digestibility (determined by the pH stat procedure and the dialysis equilibrium method) of soy protein with lower phytate content (0.07%) was greater than soy protein with a higher phytate content (1.41%) . In vitro kinetic studies were conducted with a range of soy protein and phytate concentrations to determine the velocity of the hydrolysis 21 reaction in the presence of the enzyme pronase. Results of the kinetic studies indicated that differences in the digestibility of the soy proteins were not due to stearic hindrance by phytate at enzyme-substrate reaction sites but probably due to accumulation of end products. In addition to its interactions with protein, phytate is known to have an adverse effect on mineral nutrition. Under favorable conditions, phytic acid may precipitate all polyvalent cations. The insolubility of metal-phytate complexes has been. claimed to interfere with the bioavailability of calcium, zinc, and iron in humans (Morris, 1985; Graf and Dintzis, 1982; Graf and Eaton, 1984). Phytate also forms complexes with more than one mineral in a complex, eg. calcium-zinc-phytate complex. Due to the higher affinity of zinc than of calcium for phytate, calcium exhibits a bimodal effect on the solubility of zinc, i.e. low concentrations of calcium increase the solubility of zinc whereas high concentrations potentiate the precipitation of zinc by phytate (Graf and Eaton, 1984). Interactions with Iron In mature plant seeds, the inositol phosphates occur mainly as hexaphosphate, but during food processes involving prolonged heat treatment, it is likely that lower forms of inositol phosphates are formed. All inositol phosphates from 22 di- to hexaphosphates were found to form insoluble iron complexes (de Boland et al., 1975). Complete hydrolysis of inositol hexa- and pentaphosphates through activation of endogenous phytase led to a strong increase in the availability of iron estimated in vitro (Sandberg and Svanberg, 1991) . Presence of low inositol phosphates was found to induce an increase in iron solubility through the formation of small soluble iron complexes. During digestion of phytate in the stomach and small intestine it is likely that phytate is degraded to lower inositol phosphate forms (Sandberg et al., 1987). The beneficial effects of partial phytate removal and heat on iron bioavailability from soy protein-based diets, studied by chick hemoglobin repletion, were reported by Rodriguez et al. (1985). The authors attributed the beneficial effects to promotion of protein-iron-phytate complex digestion and the release of endogenous or added iron. Recently, a protein- related moiety in the conglycinin fraction of soybean protein isolate has been demonstrated to be one of the major inhibitors of iron absorption in.humans (Lynch et al., 1994). Although it is known that phytate modifies iron dialyzability/bioavailability (Brune et al., 1992; Sandberg and Svanberg, 1991; Sandberg et al., 1989; Hallberg et al., 1987 ) , it is difficult to identify a quantitative relationship between phytate content and iron dialyzability (Lombardi- Boccia et al., 1991). 23 Interactions with Zinc Although the soluble zinc in soybean apparently is not associated with phytate, phytic acid is thought to be the primary inhibitory factor in soybean products that results in reduced zinc bioavailability. A reduction in phytate is reported to improve the bioavailability of soybean zinc (Ellis and Morris, 1981; Forbes et al., 1979; Lonnerdal et al. , 1988; Zhou et al., 1992). Lei et al. (1993) reported that supplementation of corn-soybean meal diets with microbial phytase improved the bioavailability of zinc to weanling pigs. Some of the earliest experimental evidence that a zinc- phytate-protein complex was responsible for the increased requirement of zinc in an animal diet containing soybeans compared to a diet not containing soybeans was provided by O'Dell and Savage (1960). Allred et al. (1964) found that soybean protein from which phytic acid had been removed was no longer capable of binding zinc in vitro. Studies also show that zinc in particular is poorly utilized from soy products as compared to other minerals, although the presence of soy protein per se has little effect on the bioavailability of zinc from other sources in the diet (Forbes et al., 1979 ; Forbes and Parker, 1977). Changes in protein phytate interactions during processing may influence zinc bioavailability. Studies show that zinc utilization from certain soy protein isolates is much lower than from soybean meals (Rackis and Anderson, 1977) . One 24 explanation may be that isoelectric precipitation used in the manufacture of soy protein isolate causes the formation of phytate-protein complexes, which result in the formation of phytate-protein-mineral complexes in the final product, thus rendering the mineral unavailable (Cheryan, 1980). EFFECT OF EXTRUSION PROCESSING ON MINERAL BIOAVAILABILITY Chemical changes during extrusion that have been suggested to play a role in influencing mineral bioavailability are: 1) formation of Maillard products, 2) increase in lignin fraction due to heat treatment and 3) formation of amylose-lipid complexes (Fairweather-Tait et al. , 1989). An increase in in vitro iron dialyzability was reported in extruded maize based snack foods (Hazell and Johnson, 1989) and defatted soy flour (Latunde-Dada, 1991). It has been speculated that during extrusion some of the high molecular weight compounds (including phytate) are degraded, thus releasing iron (Latunde-Dada, 1991). A reduction in the phytic acid content of maize and potato after extrusion cooking was reported by Fairweather-Tait et al. (1987) and its degradation may be responsible for increases in iron availability. Some researchers have reported that extrusion decreased the amount of dialyzable iron. Lombardi-Boccia et al. (1991) found a slight decrease in iron dialyzability after extrusion cooking of legumes. This loss in iron dialyzability could be 25 due to interactions between minerals and tannins, interactions that are probably promoted by the extrusion process. Another possibility is that extrusion cooking deactivates the phytase naturally present in cereals and legumes (Kivisto et al., 1986) . Deactivation of phytase prevents degradation of phytate and thus, a decrease in the amount of dialyzable iron may be seen. Other researchers reported no changes in iron absorption in humans caused by extrusion of maize and potato products (Fairweather-Tait et al. , 1987) and high fiber cereal product (Kivistb et a1., 1986). The equipment used in the extrusion process considerably increased the iron content of products and this added iron appeared to be as available for absorption as the endogenous iron in the food. However, apparent zinc absorption decreased in humans fed extruded maize and potato products (Fairweather-Tait et al., 1987). The effect of extrusion cooking of a bran-flour mixture on body retention of iron and zinc in normal adults was demonstrated. Stable isotopes 58Fe and 67Zn were administered with nonextruded or extruded cereal with milk and isotopic retention was measured from fecal excretion. It was observed that extrusion cooking had no effect on isotope retention (Fairweather-Tait et al. , 1989) . It could be argued that the enhancement effect of milk on nonheme iron and zinc absorption (Randhawa and Kawatra, 1993) may have masked any changes in iron and zinc availability caused by extrusion. Also, the 26 conflicting results on the effects of extrusion on iron and zinc availability reported by various researchers may be partially explained by the differences in temperature, pressure, and moisture conditions used during extrusion processing. EFFECT OF EXTRUSION PROCESSING ON PHYSIOCHENICAL PROPERTIES OF PROTEINS Protein Digestibility Specific reactions during extrusion that modify the digestibility of proteins and the availability and identity of amino acids, as summarized by Phillips (1989) and Stanley (1989), are the following: 1. Disrupt plant structures - Intact plant structures represent a significant barrier to digestive enzymes, and the combination of heat and shear during extrusion may represent a very efficient way of releasing the nutrients. 2. Break/form non-covalent bonds - In practice, extrusion cooking usually produces complete denaturation of proteins measured as a reduction in solubility and enzymatic activity and ‘the formation. of extended. unfolded. protein. networks (Jeunink and Cheftel, 1979; Rhee et a1., 1981). 3. Break/form disulfide bonds - Various workers (Rhee et al., 1981; Hager, 1984) have shown that disulfide bonds contribute to the new, extended protein and networks produced by extrusion. Although reformation of extensive disulfide bonded 27 networks could prevent digestion, the low content of cysteine/cystine in plant proteins makes such an event unlikely. 4. Break/form peptide bonds - There seems to be little evidence to document a major role of peptide bond scission or formation during extrusion cooking. 5. Promote side-chain reactions with reducing sugars - Nonenzymatic browning, the Maillard reaction, is probably the most thoroughly studied side-chain reaction that degrades protein nutritional quality. This type of reaction leads to a decrease in protein quality both by lowering digestibility and by producing nonutilizable and toxic products. The chemical properties of protein are changed by the Maillard browning reaction even under mild conditions. In general, animal proteins are known to have higher Maillard browning than the plant proteins because of their'high.content of amino acids such as lysine, methionine, arginine, histidine etc. , which readily react with glucose in animal proteins (Yen et al., 1989). 6. Inactivate protease inhibitors - The destruction of protease inhibitors is seen to increase with extrusion temperature and moisture content. At constant temperature, inactivation of trypsin inhibitors in soybeans increased with product residence time and moisture content (Mustakas et al., 1970). 28 Protein Solubility Heat treatment used in extrusion processing may alter the solubility of proteins. Solubility of native proteins may be explained by the charge and hydrophobicity of protein molecules; however, solubility of heated protein solutions depends also on the molecular size which is seen to increase through.hydrophobic interactions and disulfide bond formation upon heating (Phillips, 1989). Upon heat treatment, initial denaturation of proteins may occur with little or no apparent loss in solubility; the denaturation step is usually followed by aggregation and coagulation or gelation (Nakai and Li-Chan, 1989). For salt-extractable beef muscle proteins, heating at 50°C or higher temperatures results in decreased solubility. In sharp contrast, heating of soy protein isolate solution results in increased solubility (Nakai and Li-Chan, 1989). Gujska and Khan (1991) studied the effects of high temperature extrusion on protein solubility and distribution in navy and pinto beans. SOS-PAGE showed redistribution of protein fractions in both beans. A high degree of protein insolubility was found after extrusion, which resulted in a decrease in albumin and globulin fractions and an increase in the insoluble residue. Soy Protein Texturisation The texturization of soy proteins during extrusion processing has been studied extensively because of its primary 29 importance in the manufacture of meat analogs. In native soybean protein molecules, most amino acid residues responsible for the chemical reactions during processing of soybean protein foods - such as cysteine (-SH) , cystine (S-S) , and.hydrophobic amino acid.residues.- are buried in.the inside region of the molecule, inaccessible to water. These residues become reactable with each other through the exposure from the inside by heat denaturation during extrusion processing. The unique texture of texturized soybean products produced by extrusion is the result of both the intermolecular interchange reaction between the exposed -SH and S-S groups and the intermolecular hydrophobic reaction among the exposed hydrophobic amino acid residues (Fukushima, 1991). Prudencio-Ferreira and Areas (1993) showed that disulfide linkages, and.hydrophobic and electrostatic interactions were the main stabilizing mechanisms for the 3-dimensional structure of soy protein isolate extruded at various temperature and moisture contents. Infrared spectra showed the presence of B-sheet anti-parallel structures. Peptide bonds were of negligible importance in extrusion texturization of soy protein. IHEAT PROTEINS Knowledge of the chemistry, composition and function of wheat grain proteins is needed to fully understand the effects of extrusion processing on wheat proteins. The traditional 30 classes of wheat endosperm proteins as separated by fractional extraction using the procedure of Osborne (1907) include: 1) the gluten proteins (gliadins, LMW glutenins, and residue proteins or HMW glutenins) and 2) the nongluten or soluble proteins (albumins and globulins). In addition to these classes, other types of proteins known to be present in the wheat endosperm are lipid-binding purothionins (Redman and Fisher, 1968), ligolins (Frazier, 1983), a small portion (1%) of endosperm protein associated with starch granules (Greenwell et al., 1985), the chloroform-methanol extractable "CM" proteins (Aragoncillo et a1., 1975) and the triticins (Singh and Shepherd, 1985). Gluten Proteins The traditional method of preparation of gluten involves gentle washing of a flour-water dough in an excess of water or a dilute salt solution to remove most of the starch and soluble material, until the gluten is obtained as a rubbery mass containing about 80% of the total protein of the flour (Wrigley and.Bietz, 1988). lingeneral, gliadins and.glutenins can be extracted from wheat flour based on their solubility properties; gliadins are soluble in aqueous alcohol, whereas glutenins are soluble in dilute acid, usually acetic acid. Gluten proteins, gliadins and glutenins, are generally characterized by having a high content of proline and glutamine. Overall, about one of every three amino acid 31 residues is glutamine, and about one of every seven residues is proline (Wrigley and Bietz, 1988). The low content of amino acids with charged side chains, eg. lysine, histidine, arginine, aspartate results in a low ionic character for the gluten proteins. The gliadin and glutenin fractions of gluten proteins differ slightly in amino acid composition. Gliadins have higher amounts of proline, glutamine, cysteine, isoleucine and phenylalanine whereas glutenins have higher amounts of glycine, lysine and tryptophan (Wrigley and Bietz, 1988). Gliadins and glutenins also differ in their physical properties, most notably in their viscoelasticity. Gliadin is cohesive, but with low elasticity, whereas glutenin is both cohesive and elastic. Gliadin is composed of proteins of relatively low molecular weight in comparison with the high molecular weight (HMW) proteins of the glutenin fraction. The gluten matrix is a major determinant of the unique rheological properties of wheat doughs essential to many of the food uses of wheat flour (MacRitchie et al., 1990). Nongluten or Soluble Proteins The soluble proteins of wheat, located in the embryo, aleurone layers and the endosperm account for about 20% of the total protein (Wrigley and Bietz, 1988) . The soluble proteins consist primarily of albumins and globulins and include enzymes and enzyme inhibitors involved in metabolic activity 32 (Payne and Rhodes, 1982) . Osborne ( 1907) described the extraction of albumins and globulins from wheat flour based on their solubility in water and dilute salt solution, respectively. Albumins and globulins have significantly more lysine, aspartic acid, threonine, alanine and valine than do gliadin and glutenin fractions, but have less glutamine (Wrigley and Bietz, 1988) . Because of their higher lysine contents, albumins and globulins have better nutritional values than most other wheat proteins (Lasztity, 1984). Isolation and characterization of albumins and globulins indicates great heterogeneity among the proteins; at least six different kinds of albumins and three different types of globulins were purified by Pence and Elder (1953). Although the albumin and globulin proteins are presumed to be less important to breadmaking quality than gluten proteins (Wrigley and Bietz, 1988), studies have shown that the soluble protein content is significantly correlated with protein quality and functionality of the wheat flour (Pence et (al., 1954; Campbell and Lee, 1982; Lasztity, 1984). In a review of wheat proteins and their technological significance, Schofield and Booth (1983) noted that the overall role of albumins and globulins in breadmaking is not yet clear. JUSTIFICATION Cereals and legumes are important components of a diet. Besides being a staple food in the diet of many populations, cereals provide calories, proteins and many essential minerals and vitamins. Legumes are a highly utilized crop due to the large number and diversity of genera and species. Legumes provide variety to the.diet and supply needed protein for many populations lacking animal protein (Uebersax and Songyos, 1989). Legumes are also valuable sources of ‘minerals, including iron and zinc. Extrusion processing has been identified as a key food processing technology for the future (Harper, 1989). It would prove beneficial to expand the uses of extrusion processing to include legumes and legume supplemented cereals because of their significance to people in various parts of the world. The effect of extrusion processing of cereals and legumes on mineral bioavailability is not clear; researchers have reported both beneficial and adverse effects of extrusion processing. Such conflicting data justify further investigation on this topic. The mechanism(s) by which extrusion affects mineral absorption are not known. Many mechanisms have been suggested - formation of amylose-lipid complexes, Maillard reaction products, lignin complexes. The potential for these complexes to significantly influence mineral availability is unclear. 33 34 Interactions between proteins, iron and phytate also couLd be altered during extrusion processing and influence iron availability; It. is known that proteins play an important role in the absorption of minerals by forming complexes with minerals and/or phytates thus either enhancing or inhibiting the absorption of the mineral. It is also well- documented that extrusion results in protein denaturation and changes protein digestibility and solubility. Protein denaturation caused by extrusion processing could affect protein-iron and protein-iron-phytate complex formation and determine the release of the mineral for absorption. Thus, the interactions between protein, iron and phytate as affected by extrusion processing need to be investigated. The objectives of this study were: 1. To assess the influence of different extrusion processing conditions on iron availability, phytate degradation and tannin content of legumes. 2. To evaluate the effect of extrusion processing of durum wheat and legume supplemented durum wheat on iron bioavailability using a rat model. 3. To study in vitro protein-iron-phytate and protein-zinc- phytate interactions in extruded durum wheat. 4. To determine the. effects of extrusion processing' at different temperatures on the solubility and distribution of proteins in durum wheat. 35 5. To investigate the effects of extrusion processing at different temperatures on the association of iron and phytate with the protein fractions in durum wheat; to identify the iron-binding polypeptides and determine their molecular weights. REFERENCES ALBRIGHT, K. J., GORDON, D. T., and COTTERILL, o. J. 1984. Release of iron from phosvitin by heat and food additives. J. Food Sci. 49:78-81. ALLRED, J. B., KRATZER, F. H., and PORTER, J. W. G. 1964. Some factors affecting the in ‘vitro binding of zinc by isolated soybean protein and by a-casein. Br. J. Nutr. 18:575-580. ARAGONCILLO, C., RODRIGUEZ-LOPERENA, M. A., CARBONERO, P., and GARCIA-OLMEDO, F. 1975. Nigrosine staining of wheat endosperm proteolipid patterns on starch gels. Anal. Biochem. 63:603-606. ARTHUR, C. R., and ISBISTER, J. P. 1987. Iron deficiency. Misunderstood, misdiagnosed and mistreated. Drugs. 33:171-182. BJORN-RASMUSSEN, E., and HALLBERG, L. 1979. Effect of animal proteins on the absorption of food iron in man. Nutr. Metab. 23:192-200. BOBILYA, D. J., ELLERSIECKQ M. R., GORDON, D. T., and VEUM, T. L. 1991. Bioavailabilities of zinc from nonfat dry milk, lowfat plain yogurt, and soy flour in diets fed to neonatal pigs. J. Agric. Food Chem. 39:1246-1251. BOTHWELL, T. H., CLYDESDALE, F. E., COOK, J. D., DALLMAN, P. R., HALLBERG, L., VAN CAMPEN, D., and WOLF, W. J. 1982. The effects of cereals and legumes on iron availability. The Nutrition Foundation, New York, NY. BRUNE, M., ROSSANDER-HULTEN, L., HALLBERG, L., GLEERUP, A., and SANDBERG, A.-S. 1992. Iron absorption from bread in humans: inhibiting effects of cereal fiber, phytate and inositol phosphates with different numbers of phosphate groups. J. Nutr. 122:442-449. CAMPBELL, W. P., and LEE, J. W. 1982. Structure and function in some components of the wheat grain. Lebensm. Wiss. Technol. 15:1-4. CHERYAN, M. 1980. Phytic acid interactions in food systems. CRC Crit. Rev. Food Sci. Nutr. 13:297-335. CHIDAMBARAM, M. V. , NUCCIO, M., and BATES, G. W. 1987. Nutritional iron.bioavailability and characterization of the iron binding components of pinto beans. Pages 479-490 36 37 in: Biology of Copper Complexes. J.R.J. Sorenson, ed. Humana Press, Clifton, NJ. CHUNG, M. C.-M. 1985. A specific iron stain for iron-binding proteins in polyacrylamide gels: application to transferrin and lactoferrin. Anal. Biochem. 148:498-502. CLYDESDALE, F. M. 1989. The relevance of mineral chemistry to bioavailability. Nutr. Today Mar/April:23-30. CLYDESDALE, F. M. 1988. Minerals: their chemistry and fate in food. Pages 57-94 in: Trace Minerals in Foods. K. T. Smith, ed. Marcel Dekker, Inc., New York, NY. CLYDESDALE, F. M., and NADEAU, D. B. 1984. Solubilization of iron in cereals by milk and milk fractions. Cereal Chem. 61:330-335. COLIN, M. A., TAPER, L. J., and.RITCHEY, S. J. 1983. Effect of dietary zinc and protein levels on the utilization of zinc and copper by adult males. J. Nutr. 113:1480-1488. COOK, J. D., LAYRISSE, M., MARTINEZ-TORRES, C., WALKER, R., MONSEN, E., and FINCH, C. A. 1972. Food iron absorption measured by an extrinsic tag. J. Clin. Invest. 51:805- 815. COOK, J. D., and MONSEN, E. R. 1976. Food iron absorption in human subjects III. Comparison of the effects of animal protein on nonheme iron absorption. Am. J. Clin. Nutr. 29:859-867. CRICI-ITON, R. R., PONCE-ORTIZ, Y., KOCH, M. H. J., PARFAIT, R., and STUHRMANN, H. B. 1978. Isolation and characterization of phytoferritin from pea (pisum sativum) and lentil (Lens esculenta). Biochem. J. 171:349-356. DE BOLAND, A. R., GARNER, G. B., and O'DELL, B. L. 1975. Identification and properties of "phytate" in cereal grains and oilseed products. J. Agric. Food Chem. 23:1186-1192. DEMAEYER, E. D., and ADIELS-TEGMAN, M. 1985. World Health Statist. Q., 38:302. DILOLLO, A., ALLI, I., and KERMASHA, S. 1991. Identification of phytate in proteins using polyacrylamide disc gel electrophoresis. J. Agric. Food Chem. 39:2128-2130. ELLIS, R., and MORRIS, E. R. 1981. Relation between phytic acid and trace metals in wheat bran and soy bean. Cereal Chem. 58:367-370. 38 ERDMAN, J. W., WEINGARTNER, K. E., MUSTAKAS, G. C., SCHMUTZ, R. D., PARKER, H. M., and FORBES, R. M. 1980. Zinc and magnesium. bioavailability' from. acid-precipitated and neutralized soybean protein products. J. Food Sci. 45:1193-1197. FAIBANKS, V. F., FAHEY, J. L., and BEUTLER, E. 1971. Clinical Disorders of Iron Metabolism. 2nd ed. Grune & Stratton, Inc., New York, NY. FAIRWEATHER-TAIT, S. J. 1992. Bioavailability of trace elements. Food Chem. 43:213-217. FAIRWEATHER-TAIT, S. J. , PORTWOOD, D. E. , SYMSS , L. L. , EAGLES, J. , and MINSKI , M. J. 1989 . Iron and zinc absorption in. human subjects from. a :mixed meal of extruded and nonextruded wheat bran and flour. Am. J. Clin. Nutr. 49:151-155. FAIRWEATHER-TAIT, S. J., SYMSS, L. L., SYMSS, L. L., SMITH, A. C., and JOHNSON, I. T. 1987. The effect of extrusion cooking on iron.absorption from maize and.potato. J. Sci. Food Agric. 39:341-348. FORBES, A. L., ADAMS, C. E., ARNAUD, M. J., CHICHESTER, C. 0., COOK, J. D., HARRISON, B. N., HURRELL, R. F., KAHN, S. G., MORRIS, E. R., TANNER, J. T., and WHITTAKER, P. 1989. Comparison of in vitro, animal, and clinical determinations of iron bioavailability: international nutritional anemia consultative group task force report on iron bioavailability. Am. J. Clin. Nutr. 49:225-238. FORBES, R. M., and.PARKER, H. M. 1977. Biological availability of zinc in and as influenced by whole fat soy flour in rat diets. Nutr. Rept. Int. 15:681-688. FORBES, R. M., WEINGARTNER, K. E., PARKER, H. M., BELL, R. R., and ERDMAN, J. W. 1979. Bioavailability to rats of zinc, magnesium and calcium in casein-, egg- and soy-protein containing diets. J. Nutr. 109:1652-1660. FRAZIER, P. J. 1983. Lipid-protein interactions during dough (development. Pages 189-212 in: Lipids in Cereal Technology. P.J. Barnes, ed. Academic Press, Inc., New York, NY. FUKUSHIMA, D. 1991. Recent progress of soybean protein foods: chemistry, technology, and nutrition. Food Rev. Intl. 7:323-351. 39 GRAF, E., and DINTZIS, F. R. 1982. Determination of phytic acid in foods by high-performance liquid chromatography. J. Agric. Food Chem. 30:1094-1097. GRAF, E., and EATON, J. W. 1984. Effects of phytate on mineral bioavailability in mice. J Nutr. 114:1192-1198. GREENEWELL, P., EVERS, A. D., GOUGH, B. M. and RUSSELL, P. L. 1985. Amyloglucosidase-catalysed erosion of native, surface-modified and chlorine-treated wheat starch granules. The influence of surface protein. J. Cereal Sci. 3:279-293. GREGER, J. L. 1988. Effect of variations in dietary protein, phosphorus, electrolytes, and vitamin D on calcium and zinc metabolism. Pages 205-228 in: Nutrient Interactions. C. E. Bodwell and J. W. Erdman, eds. Marcel Dekker, Inc., New York, NY. GREGER, J. L., and MULVANEY, J. 1985. Absorption of tissue distribution of zinc, iron and copper by rats fed diets containing lactalbumin, soy and supplemental sulfur- containing amino acids. J. Nutr. 115:200-210. GREGER, J. L., and SNEDEKER, S. M. 1980. Effect of dietary protein and phosphorus levels on the utilization of zinc, copper and manganese by adult males. J. Nutr. 110:2243- 2253. GUJSKA, E., and KHAN, K. 1991. High temperature extrusion effects on protein solubility and distribution in navy and pinto beans. J. Food Sci. 56:1013-1016. HAGER, D. F. 1984. Effects of extrusion upon soy concentrate solubility. J. Agric. Food Chem. 32:293-296. HALLBERG, L., and BJORN-RASMUSSEN, E. 1972. Determination of iron absorption from whole diet. A new two-pool model using two radioiron isotopes given as haem and non-haem iron. Scand. J. Haematol. 9:193-197. HALLBERG, L., BJORN-RASMUSSEN, E., HOWARD, L., and ROSSANDER, L. 1979. Dietary heme iron absorption: possible mechanisms for the absorption promoting effect of meat and for the regulation of iron absorption. Scand. J. Gastroenterol. 14:769-779. HALLBERG, L., and.ROSSANDER, L. 1982. Effect of soy protein on nonheme iron absorption in man. Am. J. Clin. Nutr. 36:514-520. HALLBERG, L., ROSSANDER, L., and SKANBERG, A.-B. 1987. 40 Phytates and the inhibitory effect of bran on iron absorption in man. Am. J. Clin. Nutr. 45:988-996. )HLLRPER, J. M. 1989. Food extruders and their applications. " Pages 1-16 in: Extrusion Cooking. C. Mercier, P. Linko and J. M. Harper, eds. Am. Assoc. Cereal Chem.: St. Paul, MN. HAZELL, T., and JOHNSON, I. T. 1989. Influence of food processing on iron availability in vitro from extruded maize-based snack foods. J. Sci. Food Agric. 46:365-374. HURRELL, R. F., LYNCH, S. R., TRINIDAD, T. P., DASSENKO, S. A., and COOK, J. D. 1988. Iron absorption in humans: bovine serum albumin compared with beef muscle and egg white. Am. J. Clin. Nutr. 47:102-107. JEUNINK, J., and CHEFTEL, J. C. 1979. Chemical and physiochemical changes in field bean and soy proteins texturized by extrusion. J. Food Sci. 44:1322-1325, 1328. KANE, A. P., and MILLER, D. D. 1984. In vitro estimation of the effects of selected proteins on iron bioavailability. Am. J. Clin. Nutr. 39:393-401. KIVISTO, B., ANDERSSON, H., CEDERBLAD, G., SANDBERG, A.-S., and SANDSTROM, B. 1986. Extrusion cooking of a high-fiber cereal product. Br. J. Nutr. 55:255-260. KOJIMA, N., WALLACE, D., and BATES, G. W. 1981. The effect of chemical agents, beverages, and spinach on the in vitro solubilization of iron from cooked pinto beans. Am. J. Clin. Nutr. 34:1392-1401. LASZTITY, R. 1984. The Chemistry of Cereal Proteins. CRC Press, Inc., Boca Raton, FL. LASZTITY, R., and LASZTITY, L. 1990. Phytic acid in cereal technology. Pages 309-371 in: Advances in Cereal Science and Technology. Vol. x. Y. Pomeranz, ed. Am. Assoc. Cereal Chem.: St. Paul, MN. LATUNDE-DADA, G. O. 1991. Some physical properties of ten soyabean varieties and effects of processing on iron levels and availability. Food Chem. 42:89-98. LEE, K. , and CLYDESDALE, F. M. 1979. Quantitative determination of the elemental, ferrous, ferric, soluble, and complexed iron in foods. J. Food Sci. 44:549-554. LEI, X., KU, P. K., MILLER, E. R., ULLREY, D. E., and YOKOYAMA, M. T. 1993. Supplemental microbial phytase 41 improves bioavailability of dietary zinc to weanling pigs. J. Nutr. 123:1117-1123. LOMBARDI-BOCCIA, G., DILULLO, G., and CARNOVALE, E. 1991. In vitro iron dialysability from legumes: influence of phytate and extrusion cooking. J. Sci. Food Agric. 599- 605. LONNERDAL, B. 1987. Protein-mineral interaction. Page 32 in: Nutrition '87. O. Levander, ed. Am. Inst. Nutr.: Bethesda, MD. LONNERDAL, B., BELL, J. G., HENDRICKX, A. G., BURNS, R. A., and KEEN, C. 1988. Effect of phytate removal on zinc absorption from soy formula. Am. J. Clin. Nutr. 48:1301- 1306. LYKKEN, G. I., MAHALKO, J., JOHNSON, R. E., MILNE, D., SANDSTEAD, H. H., GARCIA, W. J., DINTZIS, F. R., and INGLETT, G. E. 1986. Effect of browned.and unbrowned.corn products intrinsically labelled with 652n on absorption of 652n in humans. J. Nutr. 116:795-801. LYNCH, S. R., and COVELL, A. M. 1987. Iron in soybean flour is bound to phytoferritin. Am. J. Clin. Nutr. 45:866 (abstract). LYNCH, S. R., DASSENKO, S. A., MORCK, T. A., BEARD, J. L., and COOK, J. D. 1985. Soy protein products and heme iron absorption in humans. Am. J. Clin. Nutr. 41:13-20. LYNCH, s. R., DASSENKO, s. A., COOK, J. D., JUILLERAT, M.-A., and HURRELL, R. F. 1994. Inhibitory effect of a soybean- protein-related moiety on iron absorption in.humans. Am. J. Clin. Nutr. 60:567-572. MACRITCHIE, F., DU CROS, D. L., and WRIGLEY, C. W. 1990. Flour polypeptides related to wheat quality. Pages 79-145 in: Advances in Cereal Science and Technology. Vol. X. Y. Pomeranz. ed. Am. Assoc. Cereal Chem.: St. Paul, MN. MAHONEY, A. W., and HENDRICKS, D. G. 1982. Efficiency Of hemoglobin regeneration as a method of assessing iron bioavailability in food products. Pages 1-11 in: Nutritional Bioavailability of Iron. C. Kies, ed. Am. Chem. Soc.: Washington, DC. MARTINEZ-TORRES, C., ROMANO, E. L., and IAXRISSE, M. 1981. Effect of cysteine on iron absorption in man. Am. J. Clin. Nutr. 34:322-327. MAY, L., MORRIS, E. R. and ELLIS, R. 1980. Chemical identity 42 of iron in wheat by MOsssbauer spectroscopy. J. Agric. Food Chem. 28:1004-1006. MILLER, D. D., SCHRICKER, B. R., RASMUSSEN, R. R. and VAN CAMPEN, D. 1981. An in vitro method for estimation of iron availability from meals. Am. J. Clin. Nutr. 34:2248- 2256. MONSEN, E. R. 1988. Protein-iron interactions: influences on absorption, metabolism, and status. Pages 149-162 in: Nutrient Interactions. C. E. Bodwell and J. W. Erdman, eds. Marcel Dekker, Inc., New York, NY. MONSEN, E. R. 1980. Simplified method for calculating available dietary iron. Food Nutr. News, Natl. Livestock Meat Board. 51:1-4. MORRIS, E. R. 1985. Phytate and dietary mineral bioavailability. Pages 57-76 in: Phytic acid: Chemistry and Applications . E . Graf , ed . Pilatus press , Minneapolis, MN. MORRIS, E. R., and ELLIS, R. 1976. Isolation of monoferric phytate from wheat bran and its biological value as an iron source to the rat. J. Nutr. 106:753-760. MUSTAKAS, G. C., ALBRECHT, W. J., BOOKWALTER, G. N., MCGHEE, J. E., KWOLEK, W. F., and GRIFFIN, E. L. 1970. Extruder processing to improve nutritional quality, flavour and keeping quality of full-fat soy flour. Food Technol. 24:1290-1296. NAKAI, S., and.LI-CHAN, E. 1989. Effects of heating on.protein functionality. Pages 125-144 in: Protein Quality and the Effects of Processing. R. D. Phillips and J. W. Finley, eds. Marcel Dekker, Inc., New York, NY. NELSON, K. J., and POTTER, N. N. 1980. Iron availability from wheat gluten, soy isolate and casein complexes. J. Food Sci. 45:52-55. NELSON, K. J., and POTTER, N. N. 1979. Iron binding by wheat gluten, soy isolate, zein, albumen and casein. J. Food Sci. 44:104-111. O'DELL, B. L., and DeBOLAND, A. 1976. Complexation of phytate with proteins and cations in corn germ and oilseed meals. J. Agric. Food Chem. 24:804-808. O'DELL, B. L., and SAVAGE, J. E. 1960. Effect Of phytic acid on zinc availability; Proc. Soc. Exp. BiolJ Med. 103:304- 308. 43 OSBORNE, T. B. 1907. The proteins of the wheat kernel. Carnegie Inst. Wash., Publ. No. 84. PAYNE, P. I., and RHODES, A. P. 1982. Cereal storage proteins: structure and role in agriculture and food technology. Encycl. Plant Physiol. 14A:346-369. PECOUD, A., DONZEL, P., and SCHELLING, J. L. 1975. Effect of foodstuffs on the absorption of zinc sulfate. Clin. Pharmacol. Ther. 17:469-474. PENCE, J. W., and ELDER, A. H. 1953. the albumin and globulin proteins of wheat. Cereal Chem. 30:275-287. PENCE, J. W., WEINSTEIN, N. E., and MECHAM, D. K. 1954. The albumin and globulin contents of wheat flour and their relationship to protein quality. Cereal Chem. 31:303-311. PETERS, T., APT, L., and ROSS, J. F. 1971. Effect Of phosphates upon iron absorption studied in normal human subjects and in an experimental model using dialysis. Gastroent. 61:315-422. PHILLIPS, D. 1989. Effect of extrusion cooking on the nutritional quality of plant proteins. Pages 219-246 in: Protein Quality and the Effects of Processing. R. D. Phillips and J. W. Finley, eds. Marcel Dekker, Inc., New York, NY. PLATT, S. R, NADEAU, D. B., GIFFOR, S. R., and CLYDESDALE, F. M. 1987. Protective effect of milk on mineral precipitation by Na phytate. J. Food Sci. 52:240-241. PRICE, N. 0., BUNCE, G. E., and ENGEL, R. W. 1970. Copper, manganese and zinc balance in preadolescent girls. Nutr. Rep. Intl. 23:258-265. PRUDENCIO-FERREIRA, S. H., and AREAS, J. A. G. 1993. Protein- protein interactions in the extrusion of soya at various temperatures and moisture contents. J. Food Sci. 58:378- 381. RACKIS, J. J., and ANDERSON, R. L. 1977. Mineral availability in soy protein products. Food Prod. Dev. 11:38-44. RANDHAWA, R. K., and KAWATRA, B. L. 1993. Effect of dietary protein on the absorption and retention of zinc, iron, copper and manganese in pre-adolescent girls. Die Nahrung 37(4):399-407. 44 REDMAN, D. G., and FISHER, N. 1968. Fractionation and composition of purothionin and globulin components of wheat. J. Sci. Food Agric. 24:629-636. RHEE, K. C., KUO, C. K., and LUSAS, E. W. 1981. Texturization. pages 51-88 in: Protein Functionality in foods. J.P. Cherry, ed. Am. Chem. Soc.: Washington, DC. RITTER, M. A., MORR, C. V., and THOMAS, R. L. 1987. In vitro digestibility of phytate-reduced and phenolics-reduced soy protein isolates. J. Food Sci. 52:325-341. RODRIGUEZ, C. J., MORR, C. V., and KUNKEL, M. E. 1985. Effect of partial phytate removal and heat upon iron bioavailability from soy protein-based diets. J. Food Sci. 5:1072-1075. SANDBERG, A.-S., ANDERSSON, H., CARLSSON, N. G., and SANDSTROM, B. 1987. Degradation.products of bran phytate formed during digestion in the human small intestine: effect of extrusion cooking on digestibility. J. Nutr. 117:2061-2065. SANDBERG, A. -S. , CARLSSON, N. -G. , and SVANBERG, U. 1989. Effects of inositol tri-, tetra-, penta-, and hexaphosphates on in 'vitro estimation. of iron availability. J. Food Sci. 54:159-161. SANDBERG, A.-S., and SVANBERG, U. 1991. Phytate hydrolysis by phytase in cereals: effects on in vitro estimation of iron availability. J. Food Sci. 56:1330-1333. SANDSTROM, B., CEDERBLAD, A., and LONNERDAL, B. 1983. Zinc absorption from human milk, cow's milk, and infant formulas. Am. J. Dis. Child. 137:726-729. SATO, R., LEE, Y.-S., NOGUCHI, T., and NAITO, H. 1984. Iron solubility in the small intestine of rats fed egg yolk protein. Nutr. Rep. Intl. 30:1319-1326. SATO, R., NOGUCHI, T., and NAITO, H. 1985. The formation and iron-binding property of phosphopeptides in the small intestinal contents of rats fed.egg yolk.diet. Nutr. Rep. Intl. 31:245-252. SCHNEPF, M. I., and SATTERLEE, L. D. 1985. Partial characterization of an iron soy protein complex. Nutr. Rep. Intl. 31:371-380. SCHOFIELD, J. D., and BOOTH, M. R. 1983. Wheat proteins and their technological significance. Dev. Food Proteins. 2:1-65. 45 SCHOLMERICH, J., FREUDEMANN, A., KOTTGEN, E., WIEHOLTZ, H., STEIERT, B., LOHLE, E., HAUSSINGER, D., and GEROK, W. 1987. Bioavailability of zinc from zinc-histidine complexes. I Comparison with zinc sulfate in healthy men. Am. J. Clin. Nutr. 45:1480-1486. SCHRICKER, B. R., MILLER, D. D., and VAN CAMPEN, D. 1982. In vitro estimation of iron availability in meals containing soy products. J. Nutr. 112:1696-1705. SHAH, B. G., GIROUX, A., and BELONJE, B. 1977. Specifications for reduced iron as a food additive. J. Agric. Food Chem. 25:592-594. SINGH, N. K., and SHEPHERD, K. W. 1985. The structure and genetic control of a new class of disulphide-linked proteins in wheat endosperm. Theor. Appl. Genet. 71:79- 92. SKIKNE, B. S. 1988. Current concepts in iron deficiency anemia. Food Rev. Intl. 4:137-173. SMITH, O. B., and BEN-GERA, I. 1980. The application of high temperature short time extrusion cooking in the food industry. Pages 726-758 in: Food Process Engineering, Vol. I: Food Process Systems. P. Linko, Y. Mfilkki, J. Olkku and J. Larinkari. eds. Applied Science Publishers, London. STANLEY, D. ‘W. 1989. Protein :reactions during' extrusion cooking. Pages 321-342 in: Extrusion cooking. C. Mercier, P. Linko and J. M. Harper, eds. Am. Assoc. Cereal Chem.: St. Paul, MN. STOSKMAN, J. A. 1987. Iron.deficiency anemia: have we come far enough? JAMA 258:1645-1647. TABORSKY, G. 1983. Phosvitin. In: Iron IBinding Proteins Without Cofactors on Sulfur Clusters. E. C. Thiel, G. C. Eichorn and L. G. Marzilli, eds. Elsevier, New York, NY. TAYLOR, P. G., MARTINEZ-TORRES, C., ROMANO, E. L., and LAYRISSE, M. 1986. The effect of cysteine-containing peptides released during meat digestion on iron absorption in humans. Am. J. Clin. Nutr. 43:68-71. UEBERSAX, M., and SONGYOS, R. 1989. Extrusion and its potential for processing legumes. Unpublished material WAPNIR, R. A., and STIEL, L. 1986. Zinc intestinal absorption in rats: specificity of amino acids as ligands. J. Nutr. 116:2171-2179. 46 WOLF, W. R. 1982. Trace element analysis in food. Pages 427- 446 in: Clinical, Biochemical, and Nutritional Aspects of Trace Elements. Alan R. Liss, Inc., New York. WOOD, R. J., and HANSSEN, D. A. 1988. Effect of milk and lactose on zinc absorption in lactose-intolerant postmenopausal women. J. Nutr. 118:982-986. WRIGLEY, C. W., and BIETZ, J. A. 1988. Proteins and amino acids. Pages 159-275 in: Wheat: Chemistry and Technology. Vol. 1. Y. Pomeranz, ed. Am. Assoc. Cereal Chem.: St. Paul, MN. YEN, G.-C., LEE, T.-C., CHICHESTER, C. O. 1989. Effect of Maillard browning reaction on the chemical properties of various proteins. Pages 273-289 in: Protein Quality and the Effects of Processing. R. D. Phillips and J. W. Finley, eds. Marcel Dekker Inc., New York, NY. ZHOU, J. R., FORDYCE, E. J., RABOY, V., DICKINSON, D. B., WONG, M.-S., BURNS, R.A., and ERDMAN, J. w. 1992. Reduction of phytic acid in soybean products improves zinc bioavailability in rats. J. Nutr. 122:2466-2473. CHAPTER 2. THE INFLUENCE OF EXTRUSION PROCESSING ON IRON DIALYZABILITY, PHYTATES AND TANNINS IN LEGUMES 47 ABSTRACT Iron dialyzability, chemical forms of iron, tannin content, and.phytic acid degradation in boiled, low impact and high impact extruded. navy' beans, chickpeas, cowpeas and lentils were determined. In boiled legumes, the dialyzable iron was 1.2 to 2.7% of total iron. Dialyzable, soluble and ionic iron were highest in low impact extruded legume flours. Lower forms of inositol phosphates (inositol tri-, tetra- and penta-phosphates) increased to 51 to 71% of total phytate in legumes processed by boiling or extrusion compared to 21 to 33% in raw legumes. Tannin content was lower in extruded products compared to boiled or raw legumes. The influence of extrusion processing on iron dialyzability varied with processing conditions. Results suggest that the degradation of phytate and changes in tannin content may not be responsible for the increase in iron dialyzability and solubility associated with low impact extrusion processing. 48 INTRODUCTION Extrusion processing is being used increasingly to process new cereal and legume products. Extrusion is a high temperature, short time (HTST) cooking process which uses high shear at elevated pressure and temperature. Interactions among food components during extrusion processing may have a positive or negative effect on the bioavailability of nutrients, including bioavailability of iron. Bioavailability has been defined as the proportion of the total nutrient in a food, meal or diet that is utilized for normal body functions (Fairweather-Tait, 1992) . Important chemical factors affecting the bioavailability of iron in foods include the valence, solubility and degree of chelation or complex formation of the iron (Lee and Clydesdale, 1979). Measurements of the percentage of soluble or dialyzable iron in a food are commonly used in vitro techniques to assess potential iron bioavailability (Tanner and Whittaker, 1989). Several factors including phytates and tannins may contribute to the low bioavailability of iron in legumes which has been reported" Phytate has been identified as an inhibitor of iron bioavailability (Brune et al., 1992; Hallberg et al., 1987). During food processes involving heat 'treatment, naturally occurring' phytate in plants may be degraded to lower forms of inositol phosphates which have been reported to induce an increase in iron solubility through the 49 50 formation of small soluble iron complexes (Sandberg et al., 1989). Tannins, located mainly in the seed coat or testa of beans, are another factor that has been suggested to be responsible for the low bioavailability of iron in legumes (Torrance et al., 1982; Rao and Prabhavathi, 1982). Due to the significance of dietary fiber and complex carbohydrate in nutrition and health, an increased consumption of legumes is being promoted (Morrow, 1991). Because legumes also are an important plant source of iron, the purpose of the present research was to assess the influence of extrusion processing on iron bioavailability by determining in vitro iron dialyzability and chemical forms of iron. The effect of extrusion processing on phytate degradation and tannin content of legumes also was determined. MATERIALS AND METHODS Legumes Domestically grown navy beans, chickpeas, cowpeas and lentils were obtained from Morrice Grain and Bean Company, Morrice, MI. All dry bean samples were ground using a Fitzpatrick mill with a 3A 187 mesh screen to obtain legume flours. Extrusion Processing A Baker Perkins, MPF-SO (25:1 L/D) corotating twin-screw extruder was used to extrude the legume flours under low and high impact conditions. The major differences in the configuration of low and high impact processes include a) screw configuration, b) screw speed, c) moisture content and d) barrel zone temperatures. The specific conditions of the two extrusion processes are as follows: Low impact High impact Screw speed (rpm) 350 400 Added moisture (%) 16 10 - 14 Die temperature (°C) 105 - 117 131 - 137 Barrel temperature (°C) 82 - 93 101 - 108 The extruded products (moisture content - 5.8 to 13.6%) were ground using the Fitzpatrick mill with a 0.10 cm mesh screen prior to analyses. 51 52 Boiling The non-extruded legumes were cooked using a conventional home cooking procedure. The legumes were soaked overnight in tap water (bean:water - 1:3), heated.to boiling point and.held at that temperature for 30 min by which time they were soft. The boiled legumes were homogenized to a smooth consistency. All the water used for soaking or boiling the legumes was used for homogenization. The samples were stored in the refrigerator (for 3 days or less) until they were analyzed. Total Iron Analysis Extruded and non-extruded samples were wet-ashed with concentrated nitric acid and 30% hydrogen peroxide. The ash was dissolved in 0.1 N hydrochloric acid and the solution was analyzed for iron using atomic absorption spectroscopy (AAS) (Perkin-Elmer Model 2380). In Vitro Iron Dialysability The iron dialyzability assay developed by Miller et al. (1981) was used as a measure of potential iron bioavailability in boiled and extruded legume products. The extruded legume flours were mixed with water in a ratio of 1:3 (flour:water), heated to boiling temperature and held for 15 min. The slurry was cooled to room temperature prior to analysis. Iron content of the dialysate was determined using ferrozine color reagent. 53 Quantification of the Chemical Forms of Iron Boiled legumes were analyzed for elemental, total nonelemental, soluble and ionic iron according to the method proposed by Lee and Clydesdale (1979). The extruded legume flours that were prepared for the iron dialyzability assay were also analyzed for the chemical forms of iron. The complexed iron was measured as the difference between the soluble iron (determined by AAS) and ionic iron (determined using the ferrozine color reagent). Determination of Inositol Phosphates Phytic acid and its degradation products including inositol hexa-, penta-, tetra-, and tri-phosphates were determined using ion exchange chomatography and high pressure liquid chromatography techniques according to Graf and Dintzis (1982) and Sandberg and.Ahderinne (1986) modified as follows. Sample Preparation - Raw legumes were milled using a Micro- Mill (Chemical Rubber Co., OH) to pass through a sieve equipped with a 60 mesh. Samples of 0.5 g raw, boiled or extruded legume flours were extracted under mechanical agitation with 20 ml 0.5 M HCl for 2 hours at 20°C. The extract. was. centrifuged. and. supernatant. decanted, frozen overnight and filtered under pressure through a 0.47 pm membrane filter; The filtrate ‘was diluted. with 10 ml distilled deionized water (DDW) and passed through an ion exchange column containing 0.65 ml resin (AG 1-X8, 200-400 54 mesh) at 0.4 ml/min followed by 10 ml of 0.025 M HCl. Inositol phosphates were removed from the resin with ten 1 ml portions of 2 M HCl. The eluent was evaporated to dryness on a hot plate set at low temperature and diluted with 1 ml of double deionized water. Mobile Phase - The mobile phase consisted of 0.05 M formic acid:methanol (46:54) to which 1.5 ml/100 ml of tetrabutylammonium hydroxide was added. The pH was adjusted to 4.3 by addition of 9 M sulfuric acid. The mobile phase was filtered through a Millipore filter (0.45 um) under vacuum and degassed. HPLC Procedure - Inositol phosphates were separated on a reverse phase Supelcosil LC-18 column (25.0 cm x 4.6 mm) with 5 micron particle size (Supelco, Inc., PA). Injections were made with.a Rheodyne 7010 injector equipped.with.a 20 pl loop. The optimal flow rate was 1 ml/min. Inositol phosphates were detected using a differential refractometer (Waters, Model R401). Retention times and peak areas were measured with a Peak Simple II integrator (SRI Instruments, CA). Determination of Tannin Content Tannin content of legumes, expressed as catechin equivalents, was determined using a procedure described by Price et al. (1978). Ground sample (200 mg) was mixed with 10 ml of 1% concentrated HCl in methanol and incubated at room temperature for 20 min. The mixture was centrifuged at 1500 g 55 for 10 minutes and filtered through a Whatman filter # 1, and the filtrate was analyzed for tannins using the vanillin-HCl reagent. Statistical Analyses Student's t test and one-way analysis of variance followed by the test for least significant difference were used to compare means at 95% confidence level. RESULTS AND DISCUSSION Iron Dialyzability Total iron concentration was significantly higher in extruded legume flours (Table 1). A similar increase in iron content in extruded products was reported by other researchers (Fairweather-Tait et al., 1987; Lombardi-Boccia et al., 1991; Hazell and Johnson, 1989). The wear of certain parts of the extruder is presumed to be the cause for the contamination of iron in the extruded. products. However, there was no detectable elemental iron in our extruded products. Thus, the equipment used in extrusion processing may not have been responsible for the increase in iron in the extruded products. The amount of dialyzable iron in low impact extruded products was higher than in boiled legumes. However, the dialyzable iron content of high impact extruded and boiled legume products was similar except in navy beans. In navy beans, both low and high impact extruded products had higher dialyzable iron than the boiled legumes. The increase in iron dialyzability in low impact extruded products cannot be fully explained. Speculations have been made that during extrusion some of the high molecular weight compounds (including phytate) are degraded, thus releasing iron (Fairweather-Tait et al., 1987) . The reason for the differences in the effects of low and high impact extrusion processing on iron dialyzability is not readily apparent. The extent of protein denaturation during the extrusion process 56 57 Table 1. Total ilron.and.dialyzable iron.in.boiled and extruded legume products1 . Total iron Dialyzable iron (Mg/g dry wt) (Mg/9 dry wt) (% total) NAVY BEANS Boiled 61.51 1 1.25* 1.68 1 0.15* 2.74 1 0.25* Ext. low2 70.83 1 0.42* 2.85 1 0.27* 4.03 1 0.38* Ext. high3 64.71 1 1.56 2.89 1 0.34 4.47 1 0.53 CHICK PEAS Boiled 49.60 1 0.92 1.34 1 0.31 2.70 1 0.10* Ext. low2 51.86 1 0.26* 2.16 1 0.25* 4.16 1 0.48 Ext. high3 51.94 1 0.51* 1.43 1 0.24 2.76 1 0.47 COWPEAS Boiled 47.90 1 1.01* 0.65 1 0.07* 1.36 1 0.17* Ext. low2 53.87 1 0.77 1.00 1 0.08 1.86 1 0.14 Ext. high3 53.13 1 1.26* 0.87 1 0.07 1.63 1 0.14 LENTILS Boiled 57.15 1 1.52 0.68 1 0.11 1.18 1 0.20 Ext. low2 61.82 1 1.72* 1.60 1 0.21* 2.59 1 0.35* Ext. high3 58.95 1 2.09 0.70 1 0.08 1.18 1 0.14 1Each value represents the mean 1 standard deviation of samples analyzed in triplicate. 2Legume flour extruded under low impact conditions. 3Legume flour extruded under high impact conditions. *Student's t-test at 95% confidence indicates significant difference from values for respective boiled legumes. 58 and the resulting effect on protein solubility may produce variable effects on iron dialyzability. Chemical Forms of Iron Nonelemental iron in navy beans, chickpeas and cowpeas was significantly increased in the extruded products compared to their respective boiled legumes (Table 2). In lentils, an increase in nonelemental iron was seen only in the low impact extruded product compared to the boiled legume. An increase in total soluble iron in low impact extruded products compared.to the high impact extruded.or boiled legume products was observed (Table 2) . Iron solubility may be altered by extrusion processing due to the effects of heat, pressure and shear on protein solubility and distribution. Iron solubility may also be increased due to the degradation of phytate that occurs during extrusion processing. Since iron not only has to be soluble but also be in a low-molecular fomm to be dialyzable, total soluble iron (Table 2) in all legume products was higher than the dialyzable iron (Table 1) . Ionic iron was also higher in low impact extruded than high.impact extruded.or boiled.navy beans and.chickpeas (Table 2). In cowpeas and lentils, there were no significant differences in ionic iron between low'impact extruded and high impact extruded products. However, the values were higher than the ionic iron in boiled legumes. The results of the analyses of iron dialyzability and Ta ex 801 Ext Ext 59 Table 2. Nonelemental, soluble and ionic iron in boiled and extruded legume productsl. Nonelemental Soluble Ionic (pg/g dry wt) (% total) (% total) NAVY BEANS Boiled 61.51 1 1.25 7.29 1 0.15* 6.86 10.26 Ext. low2 70.83 1 0.42* 13.42 1 0.51 11.93 10.18,, Ext. high3 64.71 1 1.56* 7.33 1 1.12 5.8910.33 CHICK PEAS Boiled 49.60 1 0.92 5.33 1 1.12* 4.04 10.92 Ext. lowz 51.86 1 0.26: 14.02 1 1.00 11.73 £1.48* Ext. high3 51.94 1 0.51 5.04 1 0.54 3.43 10.27 COWPEAS Boiled 47.90 1 1.01 6.87 1 0.11 6.15 1 0.14 Ext. low2 53.87 1 0.77* 10.14 1 0.48* 7.0910.14* Ext. 1119113 53.61 1 1.20* 7.33 1 1.80 6.93 10.12* LENTILS Boiled 56.62 1 0.52 14.13 1 2.71 5.14 1 0.50 Ext. low2 59.14 1 1.12* 21.32 1 0.92* 7.5810.09* Ext. high3 56.39 1 0.67 13.80 1 0.74 7.53 1 0.19* 1Each value represents the mean 1 standard deviation of samples analyzed in triplicate. 2Legume flour extruded under low impact conditions. iLegume flour extruded under high impact conditions. Student's t-test at 95% confidence indicates significant difference from values for respective boiled legumes. tn re co pH an De< cor pre cor ext to the ext: to Cons dial degr Phosl 60 chemical forms of iron suggest that low impact extrusion processing increased iron solubility and, as a consequence, iron dialyzability. High impact extrusion, on the other hand, did not change iron solubility compared to boiled legumes, hence no differences were seen in iron dialyzability of the two legume products. Ionic iron did not seem to accurately reflect iron. dialyzabilityu The differences in the pH conditions used in the two assays may be responsible since a pH of 7.5 was used in determining iron dialyzability, whereas an acidic pH was used for the measurement of ionic iron. Degradation of Phytate The inositol tri-, tetra-, penta-, and hexaphosphate contents in raw, boiled, and extruded legume flours are presented in Table 3. Although no changes in total phytate content were observed, processing (either boiling or extrusion) increased the conversion of inositol hexaphosphate to its lower phosphate forms. Phytic acid in.raw legume flours ranged from 66 to 79% of the total inositol phosphate forms whereas, in boiled or extruded legume products, phytic acid content ranged from 20 to 50% only. Extrusion processing did not result in any consistent pattern of degradation of phytic acid. I-f iron dialyzability is directly correlated to the extent of phytate degradation, greatest conversion of phytate to lower inositol phosphates would be expected in low impact extruded legume 61 Table 3. Degradation products of phytic acid in raw, boiled, and extruded legume productsl. IP3 IP4 IP5 IP6 ----- % total phytate------- Total phytate (umole/g dry wt) NAVY BEAN Raw 0 5.4 15.4 79.3 14.0 1 1.9 Boiled 8.4 16.5 28.1 47.0 14.0 1 0.6 Ext.1ow2 9.5 21.5 37.1 37.4 15.4 1 0.1 Ext.high3 10.8 13.6 25.9 49.8 14.6 1 0.2 CHICKPEA Raw 0 13.4 20.7 66.0 7.7 1 0.1 Boiled 25.3 19.0 21.9 31.0 8.7 1 0.1 Ext.low2 25.2 25.7 25.2 24.6 6.1 1 0.4 Ext.high3 9.9 14.0 27.5 48.7 8.2 1 0.1 COWPEA Raw 0 7.1 20.8 72.3 11.5 1 0.3 Boiled 27.1 19.7 22.2 31.0 11.1 1 2.1 Ext.1ow2 14.5 24.6 29.4 31.6 9.6 1 0.1 Ext.high3 21.3 31.9 27.1 19.7 9.8 1 0.3 LENTIL Raw 0 10.8 22.8 66.6 5.2 1 0.1 Boiled 36.8 14.4 16.9 31.9 4.5 1 0.4 Ext.low2 16.9 19.7 28.2 35.4 5.2 1 0.6 Ext.high3 14.7 26.0 30.5 28.8 4.9 1 0.1 1Each value represents the mean or mean of samples analyzed in triplicate. 2Legume flour extruded under low impact conditions. 3Legume flour extruded under high impact conditions. standard deviation 62 products since iron dialyzability was highest in these products. However, this pattern was observed only in navy beans and chickpeas and not in the case of cowpeas or lentils. Previously, Sandberg et al. (1989) reported that only inositol hexa- and pentaphosphates decreased iron solubility at simulated physiological conditions and their degradation seemed to significantly reduce the inhibiting effect of phytate on iron availability. Sandberg and Svanberg (1991) reported that complete hydrolysis of inositol hexa- and pentaphosphates through activation of endogenous phytase led to a strong increase in in vitro iron availability. Our results suggest that although extrusion processing causes degradation of phytate, the presence of different forms of inositol phosphates does not appear to explain the varying effects of low impact and high impact extrusion on iron dialyzability. Tannin Content Extrusion processing decreased the tannin content of legume flours by about 31 to 76% compared to raw legumes (Table 4). The effects of low impact and high impact extrusion processing on iron dialyzability cannot be explained by the decrease in tannin content of extruded legume flours since the decrease was seen in both low impact and high impact extrusion processing. Boiling decreased tannin content by 20% in cowpeas only. No detectable tannins were seen in raw, 63 Table 4. Tannin content (mg catechin equivalents/g dry weight) of raw, boiled, and extruded legume products . Legume Raw Boiled Ext.low2 Ext.high3 Chickpea 2.3210.3 2.6410.3 1.1010.3* 1.6110.1* Cowpea 2.9110.1 2.3110.1* 1.9710.0* 1.9110.0* Lentil 7.6510.2 7.9810.8 2.2210.l* 1.8410.2* 1Each value represents the mean 1 standard deviation of samples analyzed in triplicate; navy beans showed no detectable tannins. 2Legume flour extruded under low impact conditions. iLegume flour extruded under high impact conditions. ANOVA test at 95% confidence level indicates significant difference from values for raw legumes. 64 boiled or extruded navy bean products. Boiling and extrusion processing produced different effects on the tannin content of legumes. In the preparation of boiled legumes for analyses, the water used for soaking or boiling the legumes was used for homogenization. Thus, losses of tannins due to leaching during soaking or boiling were unlikely. Another speculation is that during extrusion, tannins may complex with other components of legumes, such as proteins or sugars, and may become undetectable by the vanillin assay. In conclusion, the factors investigated in our study, phytates and tannins, do not seem to explain the varying effects of extrusion processing on iron dialyzability. Another component of legumes thought to interfere with iron bioavailability is fiber. The modification of fiber during extrusion processing is not fully elucidated; however, it has been shown that at mild or moderately severe conditions, extrusion does not significantly change dietary fiber content but solubilizes some fiber' components. .At more severe conditions, the dietary fiber content is seen to increase, mainly due to the formation of enzyme-resistant starch fractions (Asp and BjOrck, 1989; Theander and Westerlund, 1987). A.redistribution.of insoluble to soluble dietary fiber in extruded wheat flour also has been reported (BjOer et a1., 1984). The extent to which the effects of extrusion processing on iron dialyzability in legumes might be 65 attributed to alterations in the distribution or characteristics of fiber components is yet to be determined. The role of other factors such as competing minerals, lignin complexes and the interactions between the various components in determining the effects of extrusion on iron bioavailability needs to be investigated. REFERENCES ASP, N. G. and BJORCK, I. 1989. Nutritional properties of extruded foods. Pages 399-434 in: Extrusion Cooking. C. Mercier, P. Linko and J. M. Harper, eds. Am. Assoc. Cereal Chem.: St. Paul, MN. BJORCK, I., ASP, N. G., BIRKHED, D., and LUNDQUIST, I. 1984. Effects of processing on availability of starch for digestion in vitro and in vivo: I Extrusion cooking of wheat flours and starch. J. Cereal Sci. 2:91-103. BRUNE, M., ROSSANDER-HULTEN, L., HALLBERG, L., GLEERUP, A., and SANDBERG, A.-S. 1992. Iron absorption from bread in humans: inhibiting effects of cereal fiber, phytate and inositol phosphates with different numbers of phosphate groups. J. Nutr. 122:442-449. FAIRWEATHER-TAIT, S. J. 1992. Bioavailability of trace elements. Food Chem. 43:213-217. FAIRWEATHER-TAIT, S. J., SYMSS, L. L., SMITH, A. C., and JOHNSON, I.T. 1987. The effect of extrusion cooking on iron absorption from maize and potato. J. Sci. Food Agric. 39:341-348. GRAF, E., and DINTZIS, F. R. 1982. Determination of phytic acid in foods by high.performance liquid chromatography. J. Agric. Food Chem. 30:1094-1097. HALLBERG, L., ROSSANDER-HULTEN, L., and SKANBERG, A. B. 1987. Phytates and the inhibitory effect of bran on iron absorption in man. Am J. Clin. Nutr. 45:988-996. HAZELL, T., and JOHNSON, I. T. 1989. Influence of food processing on iron availability in vitro from extruded maize-based snack foods. J. Sci. Food Agric. 46:365-374. LEE, K. , and CLYDESDALE, F. M. 1979. Quantitative determination of the elemental, ferrous, ferric, soluble, and complexed iron in foods. J. Food Sci. 44:549-554. LOMBARDI-BOCCIA, G., DILULLO, G., and CARNOVALE, E. 1991. In- vitro dialysability from legumes: influence of phytate and extrusion cooking. J. Sci. Food Agric. 48:599-605. MILLER, D. D., SCHRICKER, B. R., RASMUSSEN, R. R., and VAN CAMPEN, D. 1981. An in vitro method for estimation of iron availability from meals. Am. J. Clin. Nutr. 34:2248- 2256. 66 67 MORROW, B. 1991. The rebirth of legumes. Food Technol. 45:96, 121. PRICE, M. L., SCOYOC, S. V., and BUTLER, L. G. 1978. A critical evaluation of the vanillin reaction as an assay for tannin in sorghum.¢grain. J. .Agric. Food Chem. 26:1214-1218. RAO, B. S. N., and PRABHAVATHI, T. 1982. Tannin content of foods commonly consumed in India and its influence on ionisable iron. J. Sci. Food Agric. 33:89-96. SANDBERG, A.-S., and AHDERINNE, R. 1986. HPLC method for determination of inositol tri-, tetra-, penta-, and hexaphosphates in foods and intestinal contents. J. Food Sci. 51:547-550. SANDBERG, A.-S. , CARLSSON, N.G. , and SVANBERG, U. 1989. Effects of inositol tri-, tetra-, penta-, and hexaphosphates on in vitro estimation of iron availability. J. Food Sci. 54:159-161. SANDBERG, A.-S., and SVANBERG, U. 1991. Phytate hydrolysis by phytase in cereals: effects on in vitro estimation of iron availability. J. Food Sci. 56:1330-1333. TANNER, J. T., and WHITTAKER, P. 1989. Comparison.of in vitro, animal, and clinical determinations of iron bioavailability: International Nutritional Anemia Consultative Group Task Force report on iron bioavailability. Am. J. Clin. Nutr. 49:225-238. THEANDER, O., and WESTERLUND, E. 1987. Studies on chemical modifications in heat-processed starch and wheat flour. Staerke 39:88-93. TORRANCE, J. D., GILLOY, M., MILLS, W., MAYET, F., and BOTHWELL, T. H. 1982. Vegetable polyphenols and iron absorption. Pages 819-820 in: The Biochemistry and Physiology of Iron. P. Saltman and J. Hegenauer, eds. Elsevier Biomedical, New York, NY. CHAPTER 3. BIOAVAILABILITY OF IRON IN EXTRUDED WHEAT PRODUCTS 68 ABSTRACT The effect of extrusion processing of durum wheat and durum wheat (85%)/navy bean (15%) flours on bioavailability of iron was investigated. Dialyzable iron was 24.6, 7.3 and 16.8% in durum.wheat flour, wheat pasta and wheat/bean pasta, respectively. In vivo iron bioavailability of semolina and the durum wheat pasta products was determined using a hemoglobin (Hb) repletion technique. The final Hb concentration, Hb-Fe gain and Hb regeneration efficiency (HRE) in the rats fed wheat pasta diets were significantly lower (p<0.05) than in rats fed semolina diets. However, no significant differences (p>0.05) were found between rats fed wheat/bean pasta diets and rats fed semolina diets. 69 INTRODUCTION Extrusion processing has been identified as a key food processing technique for the future (Harper, 1989) and should be expanded to include legumes and legume supplemented cereals which are of importance to people in various parts of the world. Supplementation of cereals with legumes has significant nutritional benefits including improved protein quality. Navy bean supplementation of cereals may have great potential in the manufacture of extruded products because of its mild flavor, color and the ability’ to expand. upon extrusion. However, the influence of legume supplementation of cereals on bioavailability of iron is not known. Extrusion processing is unique in its potential ability to alter iron bioavailability due to the combination of heat, pressure and shear; therefore, the purpose of this study was to evaluate the effect of extrusion processing of durum wheat flour and navy bean supplemented durum wheat flour on iron bioavailability. The bioavailability of iron in durum wheat flour and extruded durum wheat pastas was measured in vitro and compared with an in vivo study using a rat bioassay. 70 MATERIALS AND METHODS Extrusion Processing Enriched durum wheat flour and raw navy bean (Phaseolus vulgaris) flour were used to make 100% durum wheat pasta and durum wheat (85%)/navy bean ( 15%) pasta. Durum wheat flour was sized to pass completely through a 30 mesh (US standard) and contained enrichment nutrients (thiamin, riboflavin, niacin, iron and calcium) as specified by USDA. Raw navy beans were hammer milled by passing through a Fitz Mill (Model D, Comminuting Machine, The W.J. Fitzpatric Co., Chicago) equipped with 0.07 cm sieve. Mixed dry flour ingredients were fed at the rate of 3.41 kg/min into the preconditioner where the materials were partially precooked with steam and water at 99°C. The preconditioned dough was deaerated (33 cm Hg) to achieve a smooth and uniform surface in the final product. The dough was passed through a co-rotating twin-screw extruder (Model TX-80, Wenger MFG, Sabetha, KS) at a mass temperature of 205 to 210°C with a screw speed of 154 rpm. Rotatory knives were used to length cut (ca. 2.4 cm) the pastas (macaroni) which were then dried in a dryer/cooler (Series IV, Wenger MFG, Sabetha, KS) maintained at a constant temperature of 71°C to a final moisture content of 9 to 10%. In vitro Iron Dialyzability Enriched durum wheat flour, enriched semolina, wheat pasta or wheat/bean pasta were mixed with hot water (90°C) in 71 72 a 1:3 ratio and allowed to stand for 10 min. The mixture was blended to a smooth consistency and freeze-dried. The in vitro iron dialyzability assay was carried out according to the procedure of Miller et al. ( 1981) as modified by Kane and Miller (1984). Dialyzable iron is considered a measure of iron bioavailability. Ferrozine color reagent was used to determine the total iron content of dialysate solutions (Stookey, 1970). In Vivo Determination of Iron Bioavailability WM Fifty four male Sprague-Dawley weanling rats weighing 50 to 70 g (mean 62.714.6) were used. In the first 5 wk, the rats were fed an iron deficient diet containing 1.0 mg Fe/ 100 g diet. At the end of this depletion period, blood was collected from the tail artery for analysis of hemoglobin (Hb) (hemoglobin levels were 6 to 6.5 g/100 ml of blood), and body weights were recorded. The rats were divided into nine groups (six rats in each group) based on the value of hemoglobin (g/dl) times body weight (g), so that the means of hemoglobin times body weight in all groups were equal. Each group was randomly assigned to one of the nine diets described in the diet composition section. Food intake and body weights were recorded weekly. After 14 days on the test diets, the rats were anaesthetized and blood collected from the heart for analysis of hemoglobin. Iron concentration in livers was also 73 determined- All procedures for handling the rats were approved by the All-University Committee on Animal Use and Care at Michigan State University. Diet Composition The nine diets fed to rats in this study were: control diet (1.0 mg Fe/100 g diet); control diet + ferrous sulfate (2.0 mg Fe/100 g diet); control diet + ferrous sulfate (3.0 mg Fe/100 g diet); semolina diet (2.0 mg Fe/100 g diet); semolina diet (3.0 mg Fe/100 g diet); durum wheat pasta diet (2.0 mg Fe/100 g diet); durumlwheat.pasta diet (3.0 mg Fe/100 g diet); durum.wheat/bean pasta diet (2.0 mg Fe/100 g diet); and durum wheat/bean pasta diet (3.0 mg Fe/100 g diet) (Table 1). The control diet was supplemented with two levels of FeSO4.7H20 to achieve the final iron concentrations of about 2.0 and 3.0 mg/100 g diet. Uncooked semolina and prepared pastas were incorporated into animal diets following the recommendations of the American Institute of Nutrition (Bieri, 1977; Reeves et al., 1993) (Table 1). Procedures used for preparing pastas were the same as in the in vitro studies. Semolina, obtained from the same source as durum wheat flour was used due to the non-availability of a sufficient quantity of the durum wheat flour that was used to make the pastas. All.dietS‘werelanalyzed for iron.concentrations.before feeding to the animals. .m«.6 .. 360.232.. 8.22666 6:6 ..«.6 u 05% “6.2 . one new»? 6722 6.2 . x? 29.853 8.26.822 66.224 16.... u 8C .8228 16.2. u 226 >8 “6.62 n $8060 "$026 226 B 6086 0.203 336 66625 3528.282 6536228 .65.. 74 2862\menooggc 2.2 «.« 6.6 2.« 2.2 «.« 2.« «.« « 2 on 26069 8 66235 6.« 6.2 O«E...omom mm 0m momma 6.6m m.m« 288360.23 6.6m 6.6« nomad smog: 6.6m 6.6« 8:22oemm 62.6 m6.6« mm.6 mo.m« m«.2 mo.6« me.m« me.m« me.m« eonnumosoo «.2 «.2 «.2 08220.8 6.42 6.62 6.22 6.62 2.e2 «.e2 6.6« 6.6« 6.6« :2ommo «.302 «4.0%.. «.33.. 2.308 « 2 5838.5 583683 page 060:: 82.22986 652086 «280... 2.286.... 26.868 (1636 265.66 :2 63: 86625 3026 .426 862228080 .2 028.2 75 e ' al ses Diet samples were wet-ashed with concentrated HNO3 and Hfioz. The ash was dissolved in 0.1N HCl and total iron was determined using atomic absorption spectrophotometry (AAS) (Perkin-Elmer Model 2380) . Hemoglobin concentrations were determined using a cyanmethemoglobin method (hemoglobin kit from Sigma Chem. Co., Catalog No. 525-A). Rat livers were freeze-dried and wet-ashed before iron was quantitated using AAS. During all analyses, the accuracy of the ashing procedure and.the.AAS instrument was checked.by analyzing NIST (National Institute of Science and Technology) bovine liver (1577b) and NIST wheat flour (1567a). Iron content (pg/g) of NIST bovine liver was 178.513.5 (certified value: 184115). In NIST wheat flour, total iron (pg/g) was 14.710.14 (certified value: 14.510.5). 9612111512121; Hemoglobin iron (Hb-Fe) gain was calculated for each rat as the difference between Hb-Fe at the end of the repletion period and that at the start of repletion. For the calculation of initial and final Hb-Fe, blood was assumed to be 67 g/kg body weight and Hb was assumed to contain 3.35 mg Fe/g (Miller, 1982). Iron intake for each rat was calculated from food intake and the analyzed iron content of the diet. The hemoglobin regeneration efficiency (HRE) was calculated for each rat as the percentage of iron consumed that was 76 retained in circulating hemoglobin (Miller, 1982; Forbes et al., 1989). Statistical Analyses The data reported are mean values with standard error of the mean. For the in vitro data, differences between means were statistically' analyzed. by’ Student's t test at 95% confidence level. The in vivo data were analyzed using a 2- factorial ANOVA.where the types of diets and iron dosage were taken as the independent variables. When there was no significant interaction between the diet and iron dose (p>0.1), diet means of pooled data of both.doses were used for statistical comparisons. Final Hb concentration, Hb-Fe gain, HRE, and liver iron concentration were tested using the Bonferroni-Dunn's test for differences between means. RESULTS AND DISCUSSION In Vitro Results Table 2 presents the total iron and dialyzable iron contents of raw and cooked semolina, and cooked durum wheat flour, wheat pasta, and wheat/bean pasta. Iron content of durum wheat flour was similar to that of wheat pasta suggesting that iron contamination during extrusion did not occur. .Addition.of 15% raw navy bean flour slightly increased the total iron content of the pasta product compared to wheat flour. TCtal iron contents of all products are consistent with the USDA Agriculture Handbook No. 8 Food Composition Tables (USDA, 1986; USDA, 1989). Iron dialyzability was significantly lower (p<0.05) in cooked semolina (17.0%) than in cooked durum wheat flour (24.6%). Many factors could contribute to the lower dialyzability of iron in semolina compared to wheat flour. One such factor may be a greater particle size of semolina than flour' which could influence the extent of protein digestion and iron release, thus altering iron dialyzability. An increase in iron dialyzability was seen after cooking of semolina. Cooking enhances the digestibility of the product and may Change the solubility and chemical form of iron. The in vitro results suggest a loss in iron dialyzability after extrusion of durum wheat flour. Dialyzable iron in Gcooked wheat pasta (7.3%) was significantly lower (p<0.05) than in cooked durum wheat flour (24.6%) or cooked semolina (17.0%). 77 78 Table 2. Total iron and dialyzable iron (D-Fe) content (pg/g dry weight) of durumAwheat flour, semolina, wheat pasta, wheat (85%)/navy bean (15%) pasta and navy bean flour“. Product Total Fe D-Fe D-Fe (ug/g) (Hg/g) (% total) Durum wheat flour1° 32.710.55° 8.0510.39d 24.611.20d Semolinas uncooked 37.710.17Cl 2.6310.12° 7.010.32° Semolina 35 . 510 . 50°"1 6 . 0210 . 13° 17 . 010. 36° Wheat pastab 34.710.17cd 2.5310.12° 7.310.36° Wheat/bean pastab 37.011.59d 6.2010.14° 16.810.39° Navy bean flourb 67.711.15° 8.7010.61d 12.910.91f aEach value represents mean 1 SEM of triplicates; different letters within a column indicate significant differences busing Student's t test at 95% confidence level. bSamples were prepared as described in methods section and freeze-dried before analysis. 79 However, supplementation of navy bean to wheat flour produced different results. Dialyzable iron in cooked wheat/bean pasta (16.8%) was significantly higher (p<0.05) than in cooked wheat pasta (7.3%) , significantly lower (p<0.05) than in cooked wheat flour, but not significantly different (p>0.05) from cooked semolina. In Vivo Results There were no significant differences in body weight gain by anemic rats fed either control, ferrous sulfate diets or any of the test diets (Table 3). However, rats fed the highest level of iron tended to have increased body weight gains which may have been due to improved health status resulting from a faster rate of iron repletion. As expected, final Hb concentration and Hb-Fe gain were significantly higher (p<0.05) in rats fed semolina than in rats fed the low iron control diet (Tables 3 & 4). When compared to rats fed ferrous sulfate diets, the rats fed semolina diets had significantly lower (p<0.05) final Hb concentrations but no significant differences were found in Hb-Fe gain or HRE. Final Hb concentration, Hb-Fe gain, and HRE were significantly lower (p<0.05) in the rats fed wheat pasta compared to rats fed either ferrous sulfate diets or semolina diets, but significantly higher (p<0.05) than in rats fed the control diet (Table 4) . The final hemoglobin concentration, Hb-Fe gain and HRE in rats fed wheat/bean pasta 80 .3820me «83% 5.802580: I Mama .Weuwofiomuouogflflmoggfimhgflguflégfifi 83 6.2262 «.6262 6.6226 2.6262 b.2262 6.2.263 2.2.2262 p.622 6.2.38 0.2 892.2 8v .23 2.6.8.2 2.6.8.6 «.6266 «.6266 «.6266 «626.... 2.62.2 «.6266 «6.2.6.2 18.8.2.2 6.32.8 6.26.2 6.6.8.8 6.26.66 6.22:. 222.2 2.2.26.8 p.228 2.2.26.8 £421: 35 2.6.8.2 «626.6 6.622... «.6266 «.626.« 2.68.2 6.62.6 «.6222 «.6222 6280.173. 3&9 «626.22 6.62.2 «.6222 n2.6.26.2 6.6.2262 1624.6 2.63.62 6.62222 «.6266 82282.2 29s «.6222. «.6262 2.6.26.6 «.626.» «.622... «.63... «.6266 «.6266 2.6fl.« 06.88220...— as 6.262 6.22222 6.11228 6.6226 «.6232 1.222 6.623% 6.62252 >622 923:2 68.... 6.22% p.226 p.266 6.266 6.223 6.22 6.226 «.28 p.23 19 566.223 866255 2.« 6.6 2.« 2.« «.« 2.« «.« «.« «.2 0.2 “.026 «:39... «tonne « 2&me 2.308 2 68636023 060.23 32.22658 «Looms 668388.. 6.6923 322906 2-48.0.2 20.880 (0535009 820028... 5662690: on... 68.2.5 8.8 :2 3088.2 86.55.68 6:6 68.560 .26 5222226229820. 8.22 . m 032 81 Table 4. Statistical analyses of data obtained from rat study“. Diet vs Final Hb-Fe HRE [Hb] gain Semolina Control 8 S NS Peso4 S NS NS Wheat pasta Control S S S FeSO4 S s S Semolina S S S Wheat/bean pasta FeSO4 S S S Semolina NS NS NS aS - significantly different; NS - not significantly different using 2-factorial ANOVA followed by Bonferroni-Dunn's test at 95% confidence level. Since there was no interaction between diets and dose level (p>0.1), diet means over both.doses were used for statistical comparisons. 82 were significantly lower (p<0.05) than in rats fed ferrous sulfate diets; however, they were not significantly different (p>0.05) from rats fed semolina diets. ILiver iron concentrations did not show a consistent pattern. Since they represent iron stores in the animal, the repletion period may need to have been continued for a longer time to observe significant changes. The results of the in vivo study suggest that the extruded products had lower iron bioavailability than ferrous sulfate; bioavailability of iron decreased in wheat pasta and this effect of extrusion was not seen when pasta.was made from durum wheat flour supplemented with navy bean flour. Contradictory results have been reported on the influence of extrusion on iron.bioavailability. .An increase in in vitro iron dialyzability was reported in extruded maize-based snack foods (Hazell and Johnson, 1989) and defatted soy flour (Latunde-Dada, 1991) . Speculations have been made that during extrusion some of the high molecular weight compounds are degraded, thus releasing iron. Other researchers have reported that extrusion decreased iron dialyzability (Lombardi-Boccia et al., 1991; Kivisto et al., 1986) or produced no changes in iron absorption in humans (Fairweather- Tait et al., 1987; Fairweather-Tait et al., 1989). In previous experiments, we observed that, in the case of legume flours, percent dialyzable iron varies in the same products extruded under different temperature, pressure and 83 shear conditions thus indicating that extrusion conditions play a major role in determining the effect of extrusion on iron bioavailability (Ummadi et al., 1993). Differences in extrusion conditions used by different researchers may partially account for the disagreement between studies involving extruded products and suggests that further work needs to be done to fully understand the effects of extrusion on iron bioavailability. The results obtained in the rat study indicate that iron bioavailability is lower in an extruded wheat flour pasta than in non-extruded durum wheat. The extruded products also had lower iron bioavailability compared to the diets containing the standard source of iron, i.e. ferrous sulfate. The relatively higher bioavailability of iron in uncooked semolina observed in the animal study compared.to the in vitro study is not clear. The specific effects of extrusion cannot be evaluated on the basis of the results of the animal study because the non-extruded product tested (semolina) was not the same durum wheat flour that was used to make the pastas. However, the results of the in vitro study clearly demonstrate the negative effects of extrusion on iron dialyzability of durum wheat flour. Both in vitro and.rat studies showed.that the addition of 15% navy bean flour to durum wheat flour before extrusion compensated for the negative effects of extrusion processing on iron bioavailability in durum wheat products. An 84 explanation for this improvement is not readily apparent. It might be speculated, however, that a slight improvement in protein quality of the wheat/bean products may have an impact on iron bioavailability. Another explanation could be that the introduction of bean protein favorably alters the protein- phytate-iron interactions during extrusion thus increasing iron bioavailability. At present, the mechanism(s) by which extrusion may alter mineral bioavailability is poorly understood. Many mechanisms have been suggested including amylose-lipid complexes, Maillard reaction products and increase in the lignin fraction (Fairweather-Tait et al., 1989). It may also be proposed.that alterations in bioavailability of iron during extrusion may be due to interactions between protein, phytate and iron. It is well documented that high temperature extrusion (used to make our product) results in extensive protein denaturation, decreased protein digestibility and solubility (Jeunink and Cheftel, 1979; Phillips, 1989; Gujska and Khan, 1991). Protein denaturation caused by processing could enhance protein-phytate-mineral complex formation and thus alter mineral bioavailability. REFERENCES BIERI, J. G. 1977. Report of the American Institute of Nutrition Ad Hoc Committee on standards for nutritional studies. J. Nutr. 107:1340-1348. FAIRWEATHER-TAIT, S. J., SYMSS, L. L., SMITH, A. C., and JOHNSON, I. T. 1987. The effect of extrusion cooking on iron absorption from maize and potato. J. Sci. Food Agric. 39:341-348. FAIRWEATHER-TAIT, S. J. , PORTWOOD, D. E. , SYMSS, L. L. , EAGLES, J ., and MINSKI, M. J. 1989. Iron and zinc absorption in human subjects from a mixed meal of extruded and nonextruded wheat bran and flour. Am. J. Clin. Nutr. 49:151-155. FORBES, A. L., ADAMS, C. E., ARNAUD, M. J., CHICHESTER, C. O., COOK, J. D., HARRISON, B. N., HURRELL, R. F., KAHN, S. G., MORRIS, E. R., TANNER, J. T., and WHITTAKER, P. 1989. Comparison of in vitro, animal, and clinical determinations of iron bioavailability: international nutritional anemia consultative group task force report on iron bioavailability. Am. J. Clin. Nutr. 49:225-238. GUJSKA, E. , and KHAN, K. 1991. High temperature extrusion effects on protein solubility and distribution in navy and pinto beans. J. Food Sci. 56:1013-1016. HARPER, J. M. 1989. Food extruders and their applications. Pages 1-16 in: Extrusion Cooking. C. Mercier, P. Linko and J. M. Harper, eds. Am. Assoc. Cereal Chem.: St. Paul, MN. HAZELL, T., and JOHNSON, I. T. 1989. Influence of food processing on iron availability in vitro from extruded maize-based snack foods. J. Sci. Food Agric. 46:365-374. JEUNINK, J., and CHEFTEL, J. C. 1979. Chemical and physiochemical changes in field bean and soy proteins texturized by extrusion. J. Food Sci. 44:1322-1325, 1328. KANE, A. P., and MILLER, D. D. 1984. In vitro estimation of the effects of selected proteins on iron bioavailability. Am. J. Clin. Nutr. 39:393-401. KIVISTO, B., ANDERSSON, H., CEDERBLAD, G., SANDBERG, A.-S., and SANDSTROM, B. 1986. Extrusion cooking of a high-fiber cereal product. Br. J. Nutr. 55:255-260. 85 86 LATUNDE-DADA, G. O. 1991. Some physical properties of ten soyabean varieties and effects of processing on iron levels and availability. Food Chem. 42:89-98. LOMBARDI-BOCCIA, G., DILULLO, G., and CARNOVALE, E. 1991. In vitro iron dialyzability from legumes: influence of phytate and extrusion cooking. J. Sci. Food Agric. 599- 605. MILLER, J. 1982. Assessment of dietary iron availability by rat Hb repletion assay. Nutr. Rep. Int. 26:993-1005. MILLER, D. D., SCHRICKER, B. R., RASMUSSEN, R. R., and VAN CAMPEN, D. 1981. An in vitro method for estimation of iron availability from meals. Am. J. Clin. Nutr. 34:2248- 2256. PHILLIPS, D. R. 1989. Effect of extrusion cooking on the nutritional quality of plant proteins. Pages 219-246 in: Protein Quality and Effects of Processing. R. D. Phillips and J. W. Finley, eds. Marcel Dekker, Inc., New York, NY. REEVES, P. G., NIELSEN, F. H., and FAHEY, G. C. Jr. 1993. AIN- 93 purified diets for laboratory rodents: final report of the American Institute of Nutrition Ad Hoc Writing Committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 123:1939-1951. STOOKEY, L. L. 1970. Ferrozine - a new spectrophotometric reagent for iron. Anal. Chem. 42:771-774. UMMADI, P., CHENOWETH, W., UEBERSAX, M., and OCCEANA, L. 1993. Factors affecting bioavailability of iron in extruded legumes. FASEB J. 7:1777. USDA 1986. Agriculture Handbook No. 8 Series - Composition of Legume and Legume Products (8-16) . Page . . Nutrition Monitoring Division, Human Nutrition Information Service, U.S. Dept. of Agriculture, Washington, DC. USDA 1989. Agriculture Handbook No. 8 Series - Composition of Foods: Cereal Grains and Pasta (8-20) . Page 81. Nutrition Monitoring Division, Human Nutrition Information Service, U.S. Dept. of Agriculture, Washington, DC. CHAPTER 4. PROTEIN-PHYTATE-MINERAL INTERACTIONS IN EHTRUDED PRODUCTS 87 INTRODUCTION Durum wheat flour and navy bean supplemented durum wheat flour were extruded to make pasta products. The effects of extrusion on iron bioavailability from these products, determined by in vitro and in vivo assays suggest that iron availability is lower in extruded durum wheat pasta compared to non-extruded durum wheat or navy bean supplemented durum wheat pasta (Chapter 3). In order to explain the effects of extrusion and navy bean supplementation on iron dialyzability, the role of protein-phytate-iron associations were investigated in this study using in vitro assays. Since zinc balance is of concern in infants (Hambidge, 1986; Keen and Gershwin, 1990) and extrusion is widely used in the manufacture of weaning foods, it is important to study the effects of extrusion on zinc interactions with other components in a food. Since the solubility and the molecular weight of a mineral complex are considered to be important factors affecting mineral dialyzability (Clydesdale, 1989), the effects of extrusion, dephytinization and digestion/dialysis with pepsin and pancreatin on the soluble forms of iron and zinc were determined. Regular and dephytinized durum wheat flour, navy bean flour, durum wheat pasta and.wheat/navy bean pasta were analyzed for total and soluble protein, phytate, iron and zinc before and after in vitro digestion. 88 MATERIALS AND METHODS Extrusion Processing Procedures used in the manufacture of pastas made from durum.wheat flour and durum.wheat (85%)/navy bean flour (15%) have been described in Chapter 3. Cooking Durum wheat pasta, wheat/navy bean pasta, durum wheat flour and navy bean flour were cooked by mixing the product with hot water (90°C) in a 1:3 ratio (w/v) and allowing the mixture to stand for 10 min. The mixture was then blended to a smooth consistency and freeze-dried. Dephytinization Uncooked or cooked/freeze-dried durum wheat pasta, wheat/navy bean pasta, durum wheat flour and navy bean flour were dephytinized using a phytase solution containing 100 mg of phytase (Sigma Chem. Co., P-1259) dissolved in 10 m1 of double deionized water (DDW) . To each flask containing approximately 1.00 g sample, DDW was added to make a slurry. The pH of the slurry was adjusted to 5.0, 1 ml of phytase solution was added and the mixture was incubated at 55°C for 2 h (Sandberg and Svanberg, 1991). In Vitro Iron and zinc Dialyzability In vitro digestion and dialysis in the presence of 89 90 enzymes, pepsin and pancreatin was carried out according to the procedure of Miller et al. (1981) as modified by Kane and Miller (1984). The molecular weight cutoff of the dialysis bag was 6000-8000 Da. The dialysate (solution entering the dialysis bag) and retentate (solution remaining in the flask) were centrifuged at 12,000 x G for 20 min and the supernatants were analyzed for soluble protein, phytate, iron and zinc (Fig. 1). Protein Determination The total protein contents of the flours and pastas and the protein contents of the supernatants of dialyzed and undialyzed samples were determined by the Lowry method. Phytate Determination The total phytate contents of the flours and pastas and the phytate contents of the supernatants of dialyzed and undialyzed samples were determined by ion-exchange chromatography according to Graf and Dintzis (1982) . The eluent was collected and wet-ashed with nitric acid and hydrogen peroxide. The ashed solution was analyzed for phosphorus according to Fiske and Subbarow (1925). Phytate content was calculated by using a phosphorus to phytate conversion factor of 3.55. Made to s allowed t4 Centrifug 12.00014 g collected and Zinc . 91 Uncooked flour or pasta OR Dephytinized uncooked flour or pasta 1 l l l Made to slurry, pH 7.5; Cooking followed by allowed to stand 10 min m ””0 (11363003 311d dialysis . . Retentate and dialysate (125152315: geguforriggtn at centrifuged for 20 min at coilectecf’ pe 12,000xg; supernatants collected Supematants were analyzed for soluble protein (Lowry method), soluble phytate (IEC), and soluble iron and zinc (AAS) Figure 1. Flow chart for analyses of soluble protein, phytaten’ron and zinc in raw and extruded products. 92 Iron and zinc Determination Iron and zinc contents of the supernatants of dialyzed and undialyzed samples were determined using atomic absorption spectrophotometry (Perkin-Elmer Model 2380). Statistical Analyses A one-way analysis of variance followed by the test for least significant difference at a 95% confidence interval were used to test means for significant differences. RESULTS AND DISCUSSION Extruded Wheat Pasta Table 1 presents the soluble protein, phytate, iron and zinc in durum wheat flour as affected by dephytinization and extrusion. A hypothetical model of the changes in the chemistry and interactions of protein, phytate, iron and zinc is presented below. When durum flour is dephytinized: Gluten proteins are known to have >50% of amino acids as glutamine/glutamic acid and proline (MacRitchie et al. , 1990) . At pH 7.5, the protein would have a net Charge that is slightly positive such that it may bind phytate as a binary complex. Phytate as a hexaphosphate can bind wheat proteins and decrease the solubility of the protein (Lasztity and Lasztity, 1990). Dephytinizing wheat flour may release protein which may explain the increase in protein solubility from 6.9 to 9.0 mg/g (p<0.05). Also existing in durum flour are phytate- mineral complexes which may dissociate when dephytinized releasing the minerals. Hence, an increase in soluble iron (from 20.0 to 26.3 ug/g) and zinc (from 5.7 to 10.4 ug/g) in dephytinized flour (p<0.05). When durum flour is extruded: Extrusion processing via heat, shear and pressure can 93 029666 6.2262802 662 62 33526.2 6:6 26666 B 6666 .. 83.6 68 62662622. on... 6,235 “265 6626.628 65 62 86662626 820.86 2.366626% 6:6 6622646826 .6968 :2 662222266266 6288.862. . 32986 6023022665 2.920825 :2 002222226328 362060222026 .émggggggggofi.5§§ 26:68 2.4 6.2 6.6 6.6 6.6 6.6 6.42 0226 E63 6.22 6.62 «.62 6.62 6.6 6.62 6.46 2922 2263 6«6 662 6662 664 62 666 6646 6.2.6.26 22666 6.66 6.6« 6.62 6.46 6.6« 6.6 6.62 226.866 62666 E 4 .268 9 6.4 6.6 4.62 4.6 6.2 6.6 6.62 0226 E63 6.66 6.6 6.6« 4.6« 6.6 6.6« 6.«6 2662 2263 446 66 «62 62 622 626 6646 6.2.2.616 2635 6.66 6.6« 6.6 6.46 6.64 6.6 6.662 426.2666 132.6 226.8 .33 .36662626 6.8662628: .3352». 626662626 9606626225 12262266 862.222.6168 62622266 262626266 862.266.166.22 6.2622266 L268. 62286666 666 62686 fie. 62662626 6:6 8226625216266 82662660 6n 6868 .5626 2863 .586 :2 0:26 6:6 66.22 .3268 £26286 .26 662222226266 :2 8865 .2 6226.2. 95 ' denature gluten proteins. As heat treatment continues, denaturation of proteins is followed by gelation or aggregation (Nakai and Li-Chan, 1989) leading to a decrease in solubility of proteins. However, extrusion is also known to degrade phytates to lower inositol phosphate forms (Chapter 2) which may bind to insoluble peptides and form soluble protein-phytate complexes. This may offset the decrease in protein solubility caused by extrusion. Hence, in our experiments, we did not see a significant decrease from 6.9 to 5.6 (p>0.05) in soluble protein after extrusion of durum flour. An increase in soluble phytate from 617 to 900 ug/g (p<0.05) may have been due to the degradation of inositol hexa-phosphate to its soluble, lower phosphate forms. When extruded durum pasta is dephytinized: Our results indicate that dephytinization of extruded products produced a two fold increase in protein solubility from 5.6 to 15.8 mg/g (p<0.05). It may be that extrusion processing enhances protein-phytate interactions forming either binary complexes or ternary complexes with iron or zinc. Speculations on the mechanisms by which these interactions may be promoted in the extruder are: denaturation of proteins which opens up binding sites for phytates and minerals, peptide formation and phytate degradation. When 96 extruded pasta is dephytinized, phytate is degraded and the protein-phytate or protein-mineral-phytate complexes may dissociate; thus explaining the increase in the solubilities of protein, phytate (900 to 1555 ug/g), iron (16.6 to 19.2 ug/g) and zinc (6.3 to 8.9 ug/g) (p<0.05). When durum flour is dialyzed: Protein is digested by enzymes pepsin and pancreatin increasing the solubility of protein from 6.9 to 100.3 mg/g (dialysate+retentate) (p<0.05). The free mineral content in the dialysed fractions of durum flour was negligible (0.54 ug/g iron and 0.22 ug/g zinc); hence, it is assumed that a large percentage of total iron and zinc exist as complexes with phytate, protein or other components of the flour. When durum flour is dephytinized and dialyzed: A decrease (p<0.05) in protein dialyzability from 46.3 to 28.6 mg/g was seen. However, no change in total soluble protein (dialysate+retentate) was observed following dephytinization. Phytate is degraded to lower inositol phosphate forms by phytase. Iron bound to phytate may be released causing a nonsignificant increase (p>0.05) in the dialyzability of iron from 7.6 to 8.8 ug/g. Dialyzability of zinc‘decreased significantly (p<0.05) from 1.8 to 0.9 ug/g but no change (p>0.05) in total solubility (1.8+3.4 vs 0.9+4.6) was observed; these changes may have occurred because of the 97 high molecular weight phytate-protein-zinc complexes. When extruded pasta is dialyzed: Protein solubility increased (p<0.05) from 5.6 to 89.3 mg/g (dialysate+retentate) because of protein digestion by enzymes, pepsin and pancreatin. However, the soluble protein in dialyzed wheat pasta (25.3 ug/g) is significantly lower (p<0.05) than in dialyzed wheat flour (46.3 ug/g) indicating the formation of insoluble protein complexes as a consequence of extrusion. Total soluble (dialysate+retentate) iron and zinc contents are significantly lower (p<0.05) in dialyzed wheat pasta compared to dialyzed wheat flour suggesting that insoluble protein-phytate-zinc/iron complexes formed during extrusion have not dissociated. When the extruded pasta is dephytinized and dialyzed: Total soluble (dialysate+retentate) protein, phytate, iron and zinc and dialyzable iron and zinc contents increased (p<0.05) compared to non-dephytinized dialyzed pasta. The phytate from the insoluble protein-phytate-zinc/iron complex may have been released making the protein-zinc/iron complex soluble (and somehow more susceptible to proteases?). Extruded Wheat/Bean Pasta In the case of wheat bean pasta (Table 2), similar patterns as 626.666 662626666 65 62 36666266 666 8666 B 6666 .. 83.6 666 62662626 65 6.62:6 265 66226266 65. 62 666662626 62.2666 666662616662de 666 662.26.686.26 .6868 62 662222266266 66668666.... 6029266 03802.95 doc—coon: 5. 6032226222296 3226822202.. .figoagggggégflflfigg 2363 6.6 «.6 6.62 2.6 4.6 6.62 6.66 0226 8366 6.62 6.6 6.64 6.2 6.6 2.46 6.66 2922 22666 66«2 666 666« 666 642 6662 66«6 6.2.6.666 835 6.662 6.44 6.6« 6.462 6.64 6.6« 6.6«« 252626 53.2 266 6.262 8 29666 9 6.62 6.6 2.«2 «.4 2.« «.6 6.62 0226 8:63 6.62 6.62 6.2« 6.62 6.6 6.62 2.66 2922 8366 466 46 22 624 262 666« 6666 2.626426 296.66 6.66 6.66 2.62 6.66 6.66 2.6 6.«62 4266.296 4.26622 226 .2665 .36666266 6666662626 16666262666 .666666266 6626662626 16666266666 Ag 862422662168 62622266 2226266 22662226622666.6226 6.262228 624.28. 626668666 666 626666 26.26 62662626 666 26266625666666 66 66668 .6626 6666 6>66 666 6.3.666 6666 66663 62 0:26 666 6022 .366652 .623vo 66 662222266266 62 86665 .« 6266.2. 99 in wheat pasta were observed. However, since a new type of protein (navy bean protein) was introduced along with wheat protein, the results are different. Soluble protein and phytate in undialyzed wheat bean pasta were higher (p<0.05) than in undialyzed wheat pasta. Total soluble (dialysate+retentate) protein, iron and zinc contents were also higher (p<0.05) in non-dephytinized wheat bean pasta compared to non-dephytinized wheat pasta. _Addition of navy bean to durum wheat pasta seems to have decreased the negative effects of extrusion processing on iron and zinc dialyzability and solubility. Dephytinization of wheat bean pasta increased (p<0.05) total soluble iron and zinc in the undialyzed and dialyzed samples. In conclusion, extrusion of durum wheat flour decreased soluble jprotein. and. iron. but increased soluble (phytate. Extrusion combined with dephytinization caused a significant increase in.jprotein. phytate, iron. and zinc solubility. Extruded durum wheat pasta had lower dialyzable protein, iron and zinc compared to durum wheat flour. However, dephytinization of extruded durum 'wheat pasta increased dialyzable iron and zinc compared to non-dephytinized extruded durum wheat pasta. It is speculated that high molecular ‘weight insoluble protein-phytate-iron/zinc complexes may form during extrusion and release the minerals upon phytate degradation. REFERENCES CLYDESDALE, F. M. 1989. The relevance of mineral chemistry to bioavailability. Nutr. Today Mar/April:23-30. FISKE, C. H. and SUBBAROW, Y. 1925. The colorimetric determination of phosphorus. J. Biol. Chem. 66:375-380. GRAF, E., and DINTZIS, F. R. 1982. Determination of phytic acid in foods by high-performance liquid chromatography. J. Agric. Food Chem. 30:1094-1097. HAMBIDGE, K. M. 1986. Zinc deficiency in the weanling-how important? Acta Paediatr. Scand., Suppl. 32:52-58. KANE, A. P., and MILLER, D. D. 1984. In vitro estimation of the effects of selected proteins on iron bioavailability. Am. J. Clin. Nutr. 39:393-401. KEEN, C. L., and GERSHWIN, M. E. 1990. Zinc deficiency and immune function. Ann. Rev. Nutr. 10:415-431. LASZTITY, R., and LASZTITY, L. 1990. Phytic acid in cereal technology. Pages 309-371 in: Advances in Cereal Science and Technology, Vol 10. Y. Pomeranz, ed. Am. Assoc. Cereal Chem.: St. Paul, MN. MACRITCHIE, F., DU CROS, D. L., andeRIGLEY, C. W. 1990. Flour polypeptides related to wheat quality. Pages 79-145 in: Advances in Cereal Science and Technology. Vol. x. Y. Pomeranz. ed. Am. Assoc. Cereal Chem.: St. Paul, MN. MILLER, D. D., SCHRICKER, B. R., RASMUSSEN, R. R., and VAN CAMPEN, D. 1981. An in vitro method for estimation of iron availability from meals. Am. J. Clin. Nutr. 34:2248- 2256. NAKAI, S., and LI-CHAN, E. 1989. Effects of heating on protein functionality. Pages 125-144 in: Protein Quality and the Effects of Processing. R. D. Phillips and J. W. Finley, eds. Marcel Dekker, Inc., New York, NY. SANDBERG, A.-S., and SVANBERG, U. 1991. Phytate hydrolysis by phytase in cereals: effects on in vitro estimation of iron availability. J. Food Sci. 56:1330-1333. 100 CHAPTER 5. EHTRUSION PROCESSING OF SEMOLINA. I. CHANGES IN THE SOLUBILITY AND DISTRIBUTION OF PROTEINS 101 ABSTRACT The effects of extrusion processing on protein solubility and molecular weight distribution were investigated. Enriched semolina was extruded using a twin-screw extruder at two different temperatures, 50°C and 96°C. A modified Osborne fractionation showed that the amount of protein extracted as albumin, globulin, gliadin and glutenin fractions decreased with an approximate three-fold increase in the insoluble residue following extrusion. The effect on protein solubility was greater with the higher extrusion temperature. Sodium dodecyl sulfate - polyacrylamide gel electrophoresis (SDS- PAGE) patterns of reduced and unreduced protein fractions suggest that disulfide linkages formed during extrusion may be responsible for the increase in insoluble residue fraction. 102 INTRODUCTION Wheat is the largest U.S. cereal crop used for human food (Potter, 1986). Extrusion processing of wheat is widely used in the manufacture of breakfast cereals, infant foods, crispbread, snacks and sweets and pasta products. Extrusion processing has the potential to alter protein structure, solubility and digestibility due to heat, pressure and shear (Phillips, 1989). The effects of extrusion processing on protein solubility in soybean (Cumming et al., 1973), field bean (Jeunink and Cheftel, 1979) and navy and pinto beans (Gujska and Khan, 1991) have been reported. However, little is known about the Changes in solubility and distribution of wheat proteins during extrusion. .Although proteins are known to undergo changes due to heat treatment, depending on the temperature conditions, information on hOW'the physiochemical properties of wheat proteins are affected by extrusion is not available. The purpose of this study was toldetermine the effects of extrusion processing at two different temperatures on the solubility and distribution of proteins in semolina. The role of disulfide linkages in the effects of extrusion on protein solubility and distribution also was investigated. 103 MATERIALS AND METHODS Extrusion Processing Semolina (30 mesh), milled from durum wheat with approximately 65% extraction and enriched with niacin, iron, thiamin and riboflavin was obtained from the North Dakota Mill & Elevator, Grand Forks, ND. Enriched semolina is known to have added amounts of 3.13 mg iron, 0.53 mg thiamin, 0.49 mg riboflavin. and 2.68 mg niacin. per 100 g (USDA, 1989). Semolina was extruded to make a product in the shape of small "0's". A corotating twin-screw extruder (Creusot-Loire, Model 45) was used at two different temperatures (47-50°C and 92- 96°C). Water was injected into the feed at the rate of 0.12 L/min. The screw was operated at 900 rpm and feeder was set to deliver 2.06 kg semolina/min. The final product was dried in a vat dryer/blower to a moisture content of 8 to 9%. The extruded semolina cereals were milled in a coffee mill to a coarse powder (able to pass completely through a 30 mesh sieve) prior to analyses. Protein Fractionation and Determination Albumins, globulins, gliadins, glutenins and the insoluble residue were fractionated from raw semolina and extruded semolina cereals according to the Osborne procedure as modified by Chen and Bushuk (1970). A dialysis bag with a molecular weight cutoff of 6000-8000 Da was used in the fractionation of albumins and globulins. The protein 104 105 fractions were freeze-dried and stored in a dessicator at room temperature prior to all analyses. The protein content of raw and extruded semolina and each of the protein fractions was determined by the micro-Kjeldahl method (AOAC, 1980) using a nitrogen to protein conversion factor of 5.83. Electrophoresis The protein fractions from raw semolina and extruded semolina were electrophoresed with or without a reducing agent (5% 2-mercaptoethanol) on 17.5% (w/v) gels in the presence of sodium dodecyl sulfate (SOS-PAGE) according to Ng and Bushuk (1987). The weights of protein fractions were adjusted so that the same amount of protein was loaded into each column. Gels were stained for protein using Brilliant Blue-R. Protein extract from the flour of cultivar Neepawa was used as a reference for molecular weights. Statistics Means of the percentage of total protein in each fraction were compared for statistical significance using 2-way analysis of variance followed by the test for least significant difference at 95% confidence level. RESULTS AND DISCUSSION Protein Distribution Results of the protein fractionation of raw semolina, semolina cereal extruded at 50°C (SC I) and 96°C (SC II) are presented in Table 1. The recovery of samples is reported on the basis of weight and protein content. Recovery by weight was 93 to 95% and protein recovery ranged from 93 to 98%. The variability in recovery may be attributed to the loss of low molecular weight proteins during dialysis and loss of sample during analysis. In raw semolina, protein was present in the highest amount in the alcohol-extractable protein fraction (gliadin) (41.8% of total protein) followed by the insoluble glutenin fraction (27.7%), soluble glutenin (14.2%), albumin (11.7%) and globulin (3.6%). These results are in general agreement with the findings of Chen and Bushuk (1970). The distribution of proteins differed in the extruded semolina cereals compared to raw semolina (Table 1). Extrusion.processing at both experimental temperatures caused a marked decrease in the percentage of total protein present as albumin, globulin, gliadin and glutenin fractions with a corresponding increase in the insoluble residue. The insoluble residue in raw semolina is considered to represent insoluble glutenins (Wrigley and Bietz, 1988); however, after extrusion other proteins of wheat were also extracted in this fraction. Extrusion, at a higher temperature of 96°C caused 106 107 Table 1. Protein fractionation of raw and extruded semolina. Product Raw Semolina SC I1 SC II2 Protein (%) 14.2 14.4 14.9 ALBUMIN FRACTION Weight (g/10g prod.) 0.52 1.07 0.92 Protein content (%) 31.3 14.8 2.9 % total protein4 11.7“ 10.8b 2.0c GLOBULIN FRACTION Weight (g/10g prod.) 0.09 0.03 0.24 Protein content (%) 54.6 49.9 9.6 % total protein 3.6“ 1.4 0.7c GLIADIN FRACTION Weight (g/10g prod.) 0.95 0.47 0.54 Protein content (%) 62.1 22.5 9.4 % total protein 41.8“ 7.3b 3.4° GLUTENIN FRACTION Weight (g/10g prod.) 0.33 0.19 0.18 Protein content (%) 60.1 20.7 32.4 % total protein 14.2“ 2.8 4.0 INSOLUBLE FRACTION Weight (g/10g prod.) 7.38 7.59 7.73 Protein content (%) 5.3 14.1 16.1 % total protein 27.7“ 74.0b 83.2c RECOVERY Weight (%) 92.8 95.1 94.7 Protein (%)S 98.6 96.2 93.3 1Semolina cereal prepared by extrusion processing at 47- 50°C. §Semolina cereal prepared by extrusion processing at 92-96°C. 3Protein content (N X 5.83). 4% total protein = Protein (%) x fraction wt. (g) x 10 + total protein. Means (of 2 replicates) with different superscript 5letters in each row are significantly different (p<0. 05). 5Sum of % total protein in each fraction. 108 a greater increase in insoluble residue compared to the lower temperature of 50°C. Previously, Gujska and Khan (1991) reported that extrusion temperatures of 110°C to 135°C decreased the solubility of protein constituents in navy and pinto beans. A decrease in solubility of field bean proteins after extrusion was observed by Jeunink and Cheftel (1979). They concluded that non-covalent interactions between the polypeptide chains and the formation of extended unfolded protein networks may be responsible for the decrease in solubility. Cumming et al. (1973) reported that extrusion of soy meal resulted in a four-fold loss in the solubility of water-extractable proteins and a breakdown of the remaining protein into subunits. The marked decrease in the solubility of semolina proteins at an extrusion temperature as low as 50°C (Table 1) may have potential applications in the food industry. Electrophoresis The molecular weight (MW) distribution profiles obtained from. SDS-PAGE ‘under’ reduced. or ‘unreduced. conditions for albumins of raw and extruded semolina are presented in Figure 1 (lanes 2-4, 11-13). After extrusion of raw semolina, the reduced SDS-PAGE pattern of albumins showed a disappearance of the high molecular weight polypeptides (estimated.MW: 61,500) and an appearance of a low MW (LMW) polypeptide in the range of 49,800 Da. Although the patterns were similar for both 109 Figure 1. Sodium dodecyl sulfate polyacrylamide gel electrophoretic patterns of unreduced and reduced (with 2- mercaptoethanol) albumin, globulin and gliadin fractions from raw and extruded semolina; Np-protein extract from flour of cultivar Neepawa; 2,5,8,11,14,17 represent fractions from raw semolina; 3,6,9,12,15,18 represent fractions from semolina extruded at 50°C; 4,7,10,13,16,19 represent fractions from semolina extruded at 96°C; HMW-GS - high molecular weight glutenin subunits. 110 extruded semolina cereals, extrusion at 96°C resulted in bands of greater intensity than from extrusion at 50°C. The high MW (HMW) polypeptides of albumins were most likely depolymerized after extrusion at these temperatures. There were no obvious changes between the SDS-PAGE patterns of reduced and unreduced raw (lane 2 vs 11) or extruded (lane 3 vs 12, 4 vs 13) semolina. The globulin fraction of raw semolina contained five major regions of polypeptides with estimated molecular weights of 34,600, 37,000, 45,200, 49,800, and 61,500, under reduced conditions (Fig. 1; lane 5) . The presence of several fainter bands in the MW range of 34,600 to 67,500 was also observed. Extrusion temperature of 50°C and 96°C affected protein patterns similarly causing a decrease in intensity of some bands in the HMW region (49,800 to 63,300) and an increase in intensity of some bands in the LMW region (below 45,200) (lanes 5 vs 6 and 7). There were no apparent differences between the SDS-PAGE patterns of reduced and unreduced globulin fractions from raw (lane 5 vs 14) or extruded (lane 6 vs 15, 7 vs 16) semolina, although the reduced fractions (in the MW region of 35,000-45,000 Da) diffused more than the unreduced fractions. The SDS-PAGE patterns of the reduced gliadin fraction of semolina, shown in Figure 1 (lanes 8-10) , were similar before and after extrusion at 50°C or 96°C. However, a band representing a polypeptide with estimated MW of 67,500 111 appeared fainter after extrusion at 96°C compared to raw semolina or extrusion at 50°C. A similar pattern was observed with an unreduced gliadin fraction from raw and extruded semolina (lanes 17-19). In addition, a few HMW glutenin subunits appeared in the raw semolina (lane 8) but not in either of the extruded semolina cereals. These subunits most likely were depolymerized after extrusion at the temperatures used. As in the case of globulins, there were no apparent differences between the SDS-PAGE patterns of reduced and unreduced gliadin fractions from raw (lane 8 vs 17) or extruded (lane 9 vs 18, 10 vs 19) semolina, although the reduced fractions (in the MW region of 35,000-45,000 Da) diffused more than the unreduced fractions. The reduced glutenin fraction of raw semolina appeared to have six major regions of polypeptides in the MW ranges of 37,000, 45,000-50,000, 65,000, 92,400, 100,000 and 120,000 (Fig. 2; lane 2). After extrusion, disappearance of bands in the HMW region (above 65,000), an increase in intensity of the band at 65,000, and appearance of some bands in the LMW region (below 45,200) were observed (lane 2 vs 3 and 4). Extrusion at 50°C and 96°C showed similar patterns (lane 3 vs 4); however, the 65,000 band was more intense after extrusion at the higher temperature (96°C) compared to the lower temperature (50°). There were no obvious differences in the SDS-PAGE patterns of reduced versus unreduced extruded semolina cereals (lane 3 vs 9, 4 vs 10). In raw semolina, Figure 2. Sodium dodecyl sulfate polyacrylamide gel electrophoretic patterns of unreduced and reduced (with 2- mercaptoethanol) glutenin and insoluble fractions from raw and extruded semolina; Np-protein extract from flour of cultivar Neepawa; 2,5,8,11 represent fractions from raw semolina; 3,6,9,12 represent fractions from semolina extruded at 50°C; 4,7,10,13 represent fractions from semolina extruded at 96°C; HMW-GS - high molecular weight glutenin subunits. 113 however, the unreduced pattern (lane 8) did not show the HMW polypeptides (above 65,000) and some LMW polypeptides (45,000 to 50,000) that were seen in the reduced fraction (lane 2). The reduced insoluble residue of raw semolina produced an SDS-PAGE pattern with bands in the MW ranges of 45,000-50,000 and 92,000-128,000 (Fig. 2; lane 5). After extrusion, the intensity of all bands waslgreater and there was appearance of bands in the MW range of 61,000-67,500 (lane 5 vs 6 and 7). Extrusion at 96°C (lane 7) produced slightly more intense bands compared to the extrusion temperature of 50°C (lane 6), although both showed similar SDS-PAGE patterns. In the unreduced SOS-PAGE pattern of the insoluble residue of raw semolina (lane 11) , no bands were visible. The unreduced insoluble fraction of extruded semolina cereals showed only two sets of very faint bands in the MW regions of 34,600 and 65,000 Da (lane 12, 13). Results indicate that high MW polypeptides of albumins, globulins and glutenins of semolina were most likely depolymerized after extrusion at 50°C or 96°C. In contrast, the MW distribution of gliadin proteins seemed to be unaffected by extrusion at both experimental temperatures. In the case of albumin and globulin fractions, the similarity of electrophoretic patterns of reduced versus unreduced fractions