UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An in vitro estimation of relative iron availability from wheat bran Pegg, Deborah Lynn 1983

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1984_A6_7 P45.pdf [ 4.72MB ]
Metadata
JSON: 831-1.0096164.json
JSON-LD: 831-1.0096164-ld.json
RDF/XML (Pretty): 831-1.0096164-rdf.xml
RDF/JSON: 831-1.0096164-rdf.json
Turtle: 831-1.0096164-turtle.txt
N-Triples: 831-1.0096164-rdf-ntriples.txt
Original Record: 831-1.0096164-source.json
Full Text
831-1.0096164-fulltext.txt
Citation
831-1.0096164.ris

Full Text

AN IN VITRO ESTIMATION OF RELATIVE IRON AVAILABILITY FROM WHEAT BRAN By DEBORAH LYNN PEGG B . S c , The Un i v e r s i t y of Guelph, 1981 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Food Science We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December 1983 ©D e b o r a h Lynn Pegg, I983 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y a v a i l a b l e for reference and study. I further agree that permission for extensive copying of t h i s thesis f o r scholarly purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Food Science  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date December 13. 1983 i i ABSTRACT The e f f e c t s of baking, bran p a r t i c l e s i z e and other meal components on the r e l a t i v e a v a i l a b i l i t y of the endogenous i r o n of wheat bran were investigated using an i n v i t r o method. The method simulated g a s t r o i n t e s t i n a l d i g e s t i o n and measured s o l u b l e , low molecular weight i r o n as an estimate of a v a i l a b l e i r o n . The wheat bran was incorporated i n t o a muffin product t o simulate a common domestic v e h i c l e f o r f i b e r consumption. The muffins were blended to a s l u r r y , adjusted to pH 2 and incubated with pepsin. D i a l y s i s was used to adjust the pH to i n t e s t i n a l l e v e l s and di g e s t i o n was continued with the ad d i t i o n of pancreatin and b i l e extract. Then i r o n from the d i g e s t i o n mixture which had d i f f u s e d across a semipermeable membrane ( 6000 to 8000 molecular weight c u t o f f ) was qu a n t i f i e d as percent d i a l y z a b l e i r o n . I t was found that under the conditions of the i n v i t r o estimation, e s s e n t i a l l y no i r o n was a v a i l a b l e from the bran when the muffins were combined with water. When the muffins were combined with orange j u i c e there was a very s i g n i f i c a n t enhancement of i r o n a v a i l a b i l i t y . The influence of orange j u i c e was evaluated by comparing the r e l a t i v e e f f e c t s of constituent organic a c i d s . Muffins were blended to a s l u r r y with aqueous sol u t i o n s containing e i t h e r ascorbic a c i d , c i t r i c a c i d or a combination of ascorbic and c i t r i c acids i n amounts assumed to be present i n orange j u i c e . Combination with ascorbic and c i t r i c acids together showed s i g n i f i c a n t l y greater enhancement of i r o n a v a i l a b i l i t y from wheat bran than c i t r i c acid alone which produced s i g n i f i c a n t l y more d i a l y z a b l e i r o n than ascorbic acid alone. However, the increase i n a v a i l a b l e i r o n produced by the combination of constituent organic acids was only about h a l f of that /ordduced by i i i orange j u i c e . I t was a l s o found that bran p a r t i c l e s i z e had no s i g n i f i c a n t e f f e c t on r e l a t i v e i r o n a v a i l a b i l i t y under the conditions of t h i s study. As w e l l , i t was determined that there was a s i g n i f i c a n t decrease i n i r o n a v a i l a b i l i t y due to baking. F i n a l l y , the r e s u l t s of t h i s study indicated that further research i s necessary to examine the chemistry of i r o n and i r o n binding as r e l a t e d to the a v a i l a b i l i t y of i r o n from wheat bran. iv TABLE OF CONTENTS ABSTACT i i LIST OF TABLES v i LIST OF FIGURES v i i ACKNOWLEDGEMENTS v i i i I. INTRODUCTION 1 II. LITERATURE REVIEW 3 1. IRON IN THE BODY 3 2. IRON DEFICIENCY 3 3. IRON BALANCE 4 4. AVAILABILITY OF IRON 6 5. DIETARY IRON SOURCES 6 6. CHEMISTRY OF IRON 8 7. PHYSIOLOGY OF NONHEME IRON ABSORPTION 10 8. DIETARY COMPONENTS AFFECTING IRON ABSORPTION 12 9. EFFECTS OF PROCESSING ON THE AVAILABILITY OF NONHEME IRON 14 10. EFFECT OF BRAN PARTICLE SIZE ON IRON AVAILABILITY 1? 11. EFFECT OF ORANGE JUICE ON THE AVAILABILITY OF IRON 18 12. METHODS FOR THE DETERMINATION OF IRON AVAILABILITY 20 13. IN VITRO METHODS FOR THE DETERMINATION OF IRON AVAILABILITY 23 14. CONSUMPTION OF WHOLE GRAIN PRODUCTS 28 15. AVAILABILITY OF THE ENDOGENOUS IRON OF WHEAT BRAN 32 III. MATERIALS AND METHODS 37 1. BRAN 37 2. MUFFINS 38 3- MOISTURE DETERMINATION 38 V 4. IN VITRO ESTIMATION OF RELATIVE IRON AVAILABILITY 4-1 4.1 Reagents and Materials 4l 4.2 Procedure 45 4.2.1 Slurry preparation 45 4.2.2 Pepsin digestion 45 4.2.3 Titratable acidity 4? 4.2.4 Preparation of dialysis bags 47 4.2.5 Pancreatin-bile digestion 47 4.2.6 Collection of dialysate 48 4.2.7 Analysis of dialysate 48 4.2.8 Determination of nonheme iron 50 4.2.9 Calculation of percent dialyzable iron 50 5. RESEARCH DESIGN AND ANALYSIS OF DATA 51 IV. RESULTS 53 V. DISCUSSION 58 1. INTERPRETATION OF RESULTS 58 2. AVAILABILITY OF IRON FROM WHEAT BRAN 58 3. EFFECT OF ORANGE JUICE ON IRON AVAILABILITY 65 4. EFFECT OF BAKING ON IRON AVAILABILITY 73 5. EFFECT OF BRAN PARTICLE SIZE ON IRON AVAILABILITY 76 6. LIMITATIONS OF THE STUDY 77 VI. CONCLUSIONS 78 REFERENCES 80 APPENDIX A - STANDARD CURVE FOR DETERMINATION OF DIALYZABLE IRON 87 APPENDIX B - PRELIMINARY STUDIES 88 APPENDIX C - STATISTICAL ANALYSIS 89 APPENDIX D - ANALYSIS OF SAMPLES 94 v i LIST OF TABLES I. Iron absorption from different breakfast meals 13 I I . Percentage of the total constituents of wheat present in the main morphological parts 29 I I I . The mineral content of wheat and milled fractions 30 IV. Composition of total dietary fiber of a standard wheat bran by various methods 33 V. Iron content of some foods calculated from tabular values ... Jk VI. Iron absorption by human subjects from r o l l s 35 VII. Muffin recipe 39 VIII. Chemical composition of muffin dough based on tabular values. MO EC. A summary of experimental conditions 52 X. Experimental data for the determination of % dialyzable iron (% Dl) 54 XI. The effect of acid treatment on % DI from wheat bran 56 XII. The effect of baking on % DI from wheat bran 56 XIII. The effect of bran particle size on % DI from wheat bran .... 57 v i i LIST OF FIGURES 1. Schematic of the in vitro method for determination of dialyzable iron ^2 2. -General plan of the experimental procedure ^6 v i i i ACKNOWLEDGEMENTS I would li k e to thank my husband, Neil, for his endless patience, encouragement and assistance and my parents for their support. I would also l i k e to thank Dr. J. Vanderstoep for the guidance he provided; Dr. J. Richards, Dr. S. Nakai and Dr. M. Lee for their advice, and the students and technical staff of the Department of Food Science for their suggestions and help in the laboratory. Finally, I would l i k e to acknowledge the financial support provided by the Natural Sciences and Engineering Research Council of Canada. 1 I. INTRODUCTION Iron i s an essential element for man. It is incorporated into the structure of such hody compounds as hemoglobin, myoglobin, f e r r i t i n and cytochromes. These iron-containing compounds help maintain the v i t a l cellular activities of respiration and oxygen transport. To assure the presence of iron-containing compounds, the body must maintain an iron balance. The iron lost daily from the body must be replenished by absorption of iron supplied by the diet. It i s well established that the absorption of dietary nonheme iron is influenced by a variety of factors. The iron status of the individual, the composition of the meal and the chemical form of iron a l l interact to determine the amount of nonheme iron that is absorbed by the intestinal mucosa. It is recognized that iron deficiency i s a relatively common nutritional problem worldwide. Low a v a i l a b i l i t y of dietary iron i s considered to be one of the most significant factors responsible for the deficiency. A balanced and varied diet provides sufficient total iron theoretically to meet the needs of the population. However, i t appears that much of the dietary iron i s not available for absorption. An average iron absorption from foods of only lO/o i s generally accepted. Therefore, i t i s apparent that evaluation of diets for iron adequacy requires knowledge of both the amount and a v a i l a b i l i t y of the iron present. Many claims have been made for the beneficial effects of fiber in the diet. A low fiber intake has been reported to be associated with ischemic heart disease, diabetes, diverticular disease of the colon, cancer of the colon and other diseases of the gastrointestinal tract. The reported advantages of supplementation with fiber have been recognized by commercial 2 interests and many products now include the addition of dietary fiber. The inclusion of wheat bran in such products as bread, cereals, cookies and muffins is one method of adding fiber to the diet. Wheat bran contains relatively high levels of iron when compared with other foods. However, the a v a i l a b i l i t y of the endogenous iron of this fiber source i s not well understood. The bioavailability or a v a i l a b i l i t y of iron describes that portion of the total iron in a food which i s metabolizable. Various methods exist for the measurement of iron a v a i l a b i l i t y . Chemical balance techniques, animal absorption studies and human assays using radioiron have a l l been u t i l i z e d . Several in vitro methods for the determination of iron a v a i l a b i l i t y have been proposed as alternatives to human and animal studies. The purpose of this study was to estimate the relative iron a v a i l a b i l i t y of wheat bran using an in vitro method. The method involved simulated gastrointestinal digestion followed by measurement of soluble, low molecular weight iron as an indicator of available iron. The bran was incorporated into a muffin product. The effects of baking, bran particle size and combination with known absorption-enhancing compounds on the av a i l a b i l i t y of the endogenous iron of wheat bran were investigated. It was hoped that this investigation would provide further insight into some of the factors governing the ava i l a b i l i t y of the essential element, iron. It also was hoped that the results would be beneficial to an understanding of the iron nutrition of wheat bran. 3 II. LITERATURE REVIEW 1. IRON IN THE BODY Iron is an essential element for human l i f e because i t s presence in the heme molecule permits oxygen and electron transport (Beutler, 1980). The total quantity of body iron varies with weight, hemoglobin concentration, sex and size of iron stores. The adult male has an average level of approximately 50 mg of iron per kg body weight and the adult female has about 35 rog per kg body weight (Beutler, 1980). Body iron exists in two functional catergories. Essential iron comprises about 7Q& of the total body iron and the remainder exists as nonessential storage iron found mainly in the l i v e r , spleen and bone marrow as f e r r i t i n and hemosiderin. Approximately Q% of essential body iron is incorporated in hemoglobin, % in myoglobin, \0fo in intracellular heme enzymes or serving as cofactors in other enzyme systems, and small amounts as transport iron bound to the plasma protein, transferrin (Beutler, 1980).. 2. IRON DEFICIENCY It is recognized that iron deficiency is a widespread nutritional problem (Narins, I98O; Lee and Clydesdale, 1979a). Cook (1978) reported that while iron deficiency is more prevalent in developing countries, the problem i s global. For example, at least 1Q# of menstruating women in the United States are iron deficient (Cook, 19?8). Low a v a i l a b i l i t y of. dietary iron i s considered to be one of the most important factors causing iron deficiency (Lee and Clydesdale, 1979a; Monsen et a l . , 1978). Iron deficiency occurs when the iron supply in the blood i s inadequate for normal synthesis of essential iron compounds. When iron deficiency is 4 well developed, insufficient iron is avalilable for normal hemoglobin production and anemia, characterized bysroall, pale red blood c e l l s , results (Beutler, 1980). Iron deficiency is especially l i k e l y to develop in infants and children during the f i r s t two years of l i f e , in adolescent g i r l s and in pregnant and. menstruating women (Beutler, 1980). 3- IRON BALANCE To assure the presence of iron-containing compounds in the body there must be a balance between the iron absorbed and iron lost by an individual. The amount of iron that must be absorbed from food i n order to maintain iron balance i s determined by the amount excreted, the loss in menstrual flow or from hemorrhage, the demands of pregnancy and the needs related to growth (Beutler, I98O). Once iron i s absorbed into the body most is conserved by recycling. Iron excretion i s limited. Small amounts of iron are lost in the urine, feces, skin, hair, nails and sweat (Bibeau and Clydesdale, 1976; Waddell, 1974). The main control of iron balance is achieved through regulation of iron uptake by the intestinal mucosa (Morck and Cook, I98I; Beutler, 1980). Iron balance in men i s usually not a problem. It is estimated that the adult male loses about 1 mg per day of iron. From an average dietary intake of 18 mg per day, only &/o must be absorbed to replace the normal daily losses (Subcommittee on Iron, 1979; Cook, 1978). When iron deficiency occurs in men, i t almost always signifies gastrointestinal bleeding (Cook, 1978). 5 Iron balance in women is more d i f f i c u l t to maintain because the adult female has additional requirements for iron imposed by the loss of iron in menstrual blood. Blood losses due to menstruation average about 0.6 mg per day i f distributed over the month. Therefore, the average total daily iron loss in the menstruating female is about 1.4 to 1.6 mg (Subcommittee on Iron, 1979). Women ingest about 10 to 12 mg of iron daily (Narins, I98O; Cook, 1978). As a result, women must absorb 12 to 15% of their daily iron intake to maintain iron balance (Subcommittee on Iron, 1979; Cook, 1978). The daily iron requirement for women is further increased during pregnancy. When distributed over 9 months the total iron requirement for the pregnant female i s 3.5 mg per day (Subcommittee on Iron, 1979)• Many pregnant women cannot meet their iron requirements from the diet alone and must take iron supplements (Subcommittee on Iron, 1979; Cook, 1978). In infancy, childhood and adolescence,, there are increased iron requirements due to rapid growth and i t is often d i f f i c u l t to maintain a positive iron balance (Subcommittee on Iron, 1979)' The iron necessary to replace daily losses must be absorbed from the diet. In developed countries, the diet provides approximately 5 to 7 mg of iron per 1000 kcal (Beutler, I98O). Theoretically, such an intake should more than adequately meet the iron needs of the population. However, the fact that iron deficiency continues to be a problem in spite of a plentiful intake indicates that not a l l of the iron present in the diet is capable of being absorbed (Lee and Clydesdale, 1979a). In fact, i t is estimated that only 5 to 10% of dietary iron is absorbed by healthy subjects (Beutler, I98O; Lee and Clydesdale, 1979a). Therefore, i t is evident that evaluation of diets for iron adequacy requires knowledge of both the amount and the a v a i l a b i l i t y of the iron present. 6 4. AVAILABILITY OF IRON The a v a i l a b i l i t y or bioavailability of iron describes that portion of the total iron in a food which is metabolizable (Mahoney and Hendricks, 1982) . The amount of iron potentially available from food depends on a number of complex and interacting factors. Iron absorption from food is affected by the chemical form of iron in the food, the amount of iron in the meal, the iron status of the person consuming the food, the presence of other food consitiuents in the same meal and the physiological conditions that exist in the digestive tract. The bioavailability of iron has been discussed in several reviews (Morck and Cook, I98I; Cook, 1978; Bibeau and Clydesdale, 1976; Waddell, 1974) and textbooks (Kies, 1982; Beutler, I98O; Subcommittee on Iron, 1979). The various factors that influence the a v a i l a b i l i t y of dietary iron w i l l be discussed in subsequent sections. 5. DIETARY IRON SOURCES There are two chemical forms of dietary iron; heme and nonheme, and each is absorbed by a different mechanism (Morck and Cook, I98I; Monsen et a l . , 1978) . Heme, which contains iron in a porphyrin ring structure, is found in hemoglobin and myoglobin and accounts for approximately kOfo of the iron present i n animal tissues. Heme iron enters the intestinal mucosal c e l l as the intact porphyrin complex. Within the mucosal c e l l the heme i s catabolized and the iron enters the same pathways as nonheme iron (Morck and Cook, I98I; Monsen et a l . , 1978) Nonheme iron is present in foods such as vegetables, grains, fruits and eggs and accounts for the remaining 60% of iron in animal tissues 7 (Morck and Cook, 1981; Monsen et a l . , 1978). It has been determined that a common nonheme iron pool i s formed by foods ingested in the same meal. The implication of the pool concept is that the absorption of nonheme iron depends not only on the total amount of dietary nonheme iron but also on the effects of various intraluminal factors which either enhance or inhibit the a v a i l a b i l i t y of iron (Hallberg, 1980). The absorption of heme iron from the diet is high in comparison with nonheme iron. Heme iron is prevented from interacting with other components of the diet during absorption. Therefore, the a v a i l a b i l i t y of heme iron is not influenced by the composition of the meal (Morck and Cook, 1981; Cook, 1978). Heme iron provides only 5 to 10% of the total daily iron intake in developed countries. However, due to i t s high ava i l a b i l i t y , i t provides nearly one third of the iron absorbed each day from a mixed diet (Morck and Cook, I98I; Cook, 1978). The abundance of body iron stores has only a small effect on the absorption of heme iron. An individual with no iron stores may absorb 35% of heme iron when ingested as meat while a subject with adequate iron stores may absorb 15% (Cook, 1978; Monsen et a l . , 1978). The absorption of nonheme iron i s sensitive to the iron status of •the individual. As iron stores decrease more of the iron available in the digestive tract i s absorbed and once iron stores are replete, absorption f a l l s to a low level (Beutler, I98O). Absorption of nonheme iron by an individual with adequate iron stores w i l l be about 2% from a meal containing components which inhibit the a v a i l a b i l i t y of iron. An iron deficient subject may absorb as much as 20% of nonheme iron from a meal containing 8 abundant absorption enhancers (Cook, 1978; Monsen et a l . , 1978). The absorption of nonheme iron i s more affected by the total amount of iron ingested i n a meal than is heme iron. Generally, as the level of iron in the diet increases; the proportion of the iron that is absorbed decreases, but the absolute amount increases (Narins, I98O). 6. CHEMISTRY OF IRON It i s known that the chemical characteristics of iron such as valence, solubility and type of chelation influence the bioavailability of iron from foods (Nojeim and Clydesdale, I98I; Lee and Clydesdale, 1979a). +6 -2 Iron has several oxidation states ranging from Fe to Fe , depending upon the chemical environment. The f e r r i c form, Fe +"\ and +2 the ferrous form, Fe , are the only states which occur naturally in foods (Lee and Clydesdale, 1979a). Ferrous and f e r r i c ions do not occur in the free state but are hydrated as Fe(H20)£ and F e ^ O ) ^ in acidic solutions. As the pH is raised, deprotonation of the complexed water molecules occurs. Ferric ions polymerize and eventually precipitate as f e r r i c hydroxide while most ferrous iron remains soluble in neutral and alkaline solutions (Clydesdale, 1982; Lee and Clydesdale, 1979a). At pH 7 the solubility of ferrous hydroxide is about 10"^ M. Ferric hydroxide is much more insoluble and has a solubility of lO" 1^ M at neutral pH (Lee and Clydesdale, 1979a). Numerous studies have shown that ferrous iron is absorbed and utilized more ef f i c i e n t l y by man than f e r r i c iron. It has not yet been resolved whether this i s due to a selective absorption mechanism or to the greater sol u b i l i t y of ferrous iron (Nojeim and Clydesdale, I98I; Lee and Clydesdale, 1979a). 9 It has also been found that the type of chelate formed with iron influences the absorption of nonheme iron. Many common food components are effective chelating agents or ligands (Lee and Clydesdale, 1979a). The absorption of chelated iron may be enhanced or inhibited depending on the nature of the specific iron complex including the solubility, molecular weight and s t a b i l i t y of the complexes formed with iron (Clydesdale, 1982; Miller and Schricker, 1982; Monsen and Page, 1978). Enhanced iron absorption has been observed when such compounds as ascorbic acid, sugars such as fructose, and amino acids like cysteine have been present in the meal. It has been postulated that these compounds can form low molecular weight, relatively weak chelates with f e r r i c iron. These chelates prevent the precipitation of iron in the alkaline environment of the intestine thereby rendering otherwise insoluble iron available for absorption (Morck and Cook, 1981; Beutler, I98O; Lee and Clydesdale, 1979a; Monsen and Page, 1978). Ascorbic acid and reducing agents in meat have also been reported to increase iron absorption from the diet by reducing f e r r i c iron to the more soluble ferrous form (Morck and Cook, I98I; Lee and Clydesdale, 1979a). Other dietary components such as tannins, phosphates, phytates, oxalates and carbonates have been reported to reduce the absorption of nonheme iron. It has been proposed that these iron-complexing compounds reduce absorption by the formation of insoluble precipitates or tightly bound soluble complexes with f e r r i c iron at neutral pH (Morck and Cook, 1981; Monsen and Page, 1978; Bibeau and Clydesdale, 1976). 10 7. PHYSIOLOGY OF NONHEME IRON ABSORPTION Most nonheme iron in food exists in the fe r r i c state and is bound to food components (Miller and Schricker, 1982; Innis, 1981). As has been discussed, the iron must be soluble or ligand-bound to be available for absorption. There are three phases in the absorption of nonheme iron. These phases are the intraluminal, mucosal and corporeal (Innis, 198l). In the intraluminal phase, iron is released from food in a soluble form and is ionized by the acid gastric juice. The iron enters a common intraluminal nonheme iron pool (Innis, I98I; Beutler, I98O). Narins (I98O) reported that less than half of the total iron in food is released by hydrochloric acid and peptic digestion and less than a third i s ionized. The iron released into the stomach may be reduced to the ferrous state or chelated by ligands released during digestion (Miller and Schricker, 1982; Innis, I98I; Beutler, I98O). When the products of gastric digestion pass from the stomach to the duodenum, bicarbonate secreted by the pancreas begins to neutralize the stomach acid and the pH increases from about 1.5 to 7'0 (Miller and Schricker, 1982; Innis, I98I). The previously described formation of sparingly soluble hydroxides with increasing pH can now be seen to have nutritional significance. At intestinal pH most of the fe r r i c iron precipitates as f e r r i c hydroxide unless i t is prevented by a chelating agent; while most of the ferrous iron, which is not easily precipitated is s t i l l soluble (Miller and Schricker, 1982; Innis, 1981; Beutler, 1980). In humans the mucosal c e l l s of the duodenum have the greatest capacity for iron absorption (Miller and Schricker, 1982; Innis, 1981; Beutler, I98O). In the mucosal phase of absorption soluble iron i s taken up by 11 brush border receptors and either immediately transported across to the serosal side or retained within the mucosal c e l l (Innis, 1981). The chemical form of iron that enters the mucosal c e l l , the nature of the receptor sites and the iron transport system within the c e l l are unknown (Innis, 1981; Beutler, 1980). Recently, Miller and Schricker (1982) discussed their proposal for the form of iron which is absorbed by the intestinal mucosa. Their discussion is significant to this thesis because their proposal provided the rationale for the design of the in vitro method developed by Miller et a l . (1981) which was used by this author to estimate iron a v a i l a b i l i t y . Miller and Schricker (1982) assumed that absorbable iron i s present in the duodenum as a soluble low molecular weight chelate. They also assumed that other forms of iron that may be present do not contribute significantly to absorbable iron. They provided several reasons why these assumptions seemed just i f i e d . Miller and Schricker (1982) stated that: 1. Iron may be absorbed as the intact chelate or the chelate may transfer i t s iron to an acceptor on the mucosal c e l l surface. Absorption and exchange would be much more rapid with soluble forms of iron since insoluble forms would have limited contact with the mucosal c e l l surface. 2. Iron bound to large molecular weight ligands may be available but absorption would be limited to an iron transfer mechanism since large molecules are generally not absorbed intact. Large molecular weight soluble chelates of available iron would most l i k e l y involve proteins as the ligand and digestion would quickly transform them into low molecular weight chelates. 3. Polymerized iron, even when soluble, is problably not readily available ... In an earli e r study, Peters et a l . (1971) concluded that iron i s absorbed only as the free ferrous ion or as low molecular weight complexes capable of passing through a membrane by simple diffusion. 12 The f i n a l phase of nonheme iron absorption is the corporeal phase where iron i s taken up by plasma transferrin on the serosal side of the mucosal c e l l . The exact transfer mechanism is not known but i t i s believed to involve a membrane receptor (Innis, 1981). 8. DIETARY COMPONENTS AFFECTING IRON AVAILABILITY It i s well established that the a v a i l a b i l i t y of nonheme iron can be influenced by components of foods ingested concomitantly that either enhance or inhibit the absorption of iron from a meal (Morck and Cook, 1981; Rossander et a l . , 1979; Monsen et a l . , 1978). The effect of dietary components on nonheme iron a v a i l a b i l i t y i s related to the chemistry and physiology of iron absorption which has been discussed previously. Numerous studies have been carried out on the effect of various dietary factors on iron absorption and these are reviewed by Morck and Cook (1981). One study which examined the interaction of various foods on iron absorption w i l l be discussed. Rossander et a l . (1979) measured the absorption of nonheme iron from 9 Western type breakfasts in 129 subjects using extrinsic labelling with radioiron. In the study, the iron absorption from a basal breakfast consisting of coffee (150 ml), two wheat r o l l s with margarine (12 g), one with orange marmalade (10 g) and the other with cheese (15 g), was standardised against the absorption of a reference dose of ferrous ascorbate. The basal breakfast was then used as the standard against which iron absorption from other breakfast meals was compared. As shown in Table I, there was a wide variation in the bioavailability of iron in the different breakfast meals. The greatest effects were seen with tea which reduced the absorption to less than half and with 13 Table I. Iron absorption from different breakfast meals a Experiment Composition Corrected Absorption 1 Basal breakfast (coffee) 0.16 2 with orange juice 0.40 3 with boiled egg 0.19 4 with scrambled egg and bacon 0.25 5 with cornflakes and milk 0.16 6 Basal breakfast (tea) 0.07 7 with orange juice 0.21 8 with scrambled egg and bacon 0.12 9 Basal breakfast (chocolate milk) 0.11 Adapted from Rossander et a l . (1979). ^Corrected to correspond to an absorption of 0.16 mg from the basal breakfast with coffee. 14 orange juice which increased the absorption two and a half times. 9. EFFECTS OF PROCESSING ON THE AVAILABILITY OF NONHEME IRON Almost a l l foods are processed in some form before they are consumed. It seems possible that some of these processes may influence the availability of iron. Recently, Lee (1982) reviewed the effect of processing on the bioavailability and chemistry of iron in food. Heat treatment of foods is a major form of processing and i t appears to have a varying effect on iron a v a i l a b i l i t y . Theuer et a l . (1971, 1973) investigated the effects of s t e r i l i z a t i o n on the a v a i l a b i l i t y of supplemental iron added to infant formulas. In their f i r s t study, Theuer et a l . (1971) prepared experimental batches of liquid soy isolate infant formula products without added iron and with various iron salts by standard commercial techniques. The effect of processing was assessed directly with three iron salts that were also added without processing to lyophilized product made without added iron. Iron a v a i l a b i l i t y was evaluated by measuring hemoglobin response in anemic rats fed measured amounts of the lyophilized products. It was found that heat s t e r i l i z a t i o n had l i t t l e effect on the av a i l a b i l i t y of ferrous sulfate. However processing did increase the relative a v a i l a b i l i t y of f e r r i c pyrophosphate from 39 to 93;& and of sodium iron pyrophosphate from 15 to 66%. In a similar experiment, Theuer et a l . (1973) determined the relative a v a i l a b i l i t y of various iron salts when incorporated into a liquid milk-based infant formula. It was found that s t e r i l i z a t i o n increased the relative a v a i l a b i l i t y of f e r r i c pyrophosphate from 75 to 125% and of sodium iron pyrophosphate from 40 to 60%. 15 Wood et a l . ( 1 9 7 8 ) conducted experiments to determine the effects of heat and pressure processing ( 1 , 0 5 5 g/cm2 and 121°G for 1 5 min) on the bioavailability of various iron phosphate salts using the chick hemoglobin repletion technique; Heat and pressure processing resulted in a significant enhancement of the relative biological value (RBV) of sodium f e r r i c pyrophosphate from 14 to 66% and of f e r r i c pyrophosphate from 7 to $0%. Processing did not significantly alter the RBV of ferrous sulfate or f e r r i c orthophosphate. Clemens and Mercurio ( 1 9 8 1 ) determined the av a i l a b i l i t y of carbonyl iron, electrolytic iron and f e r r i c orthophosphate in a canned cocoa-containing milk-based product. The relative biological values were determined by hemoglobin repletion assay in rats before and after retort processing. It was found that there was a three fold increase in the RBV of both carbonyl iron (from 6 5 to 202%) and electrolytic iron (from 123 to 297%) following retort processing. The RBV of f e r r i c orthophosphate remained unchanged. Disler et a l . ( 1 9 7 5 ) showed that when sugar f o r t i f i e d with ascorbic acid and f e r r i c orthophosphate was added to corn meal porridge before cooking there was a several-fold enhancement in the avai l a b i l i t y of iron from f e r r i c orthophosphate than i f i t was added after heating. However, the effect of heating f e r r i c orthophosphate alone, without ascorbate, was not determined. F r i t z and Pla ( 1 9 7 2 ) used the hemoglobin repletion assay in both rats and chicks to compare the bioavailability of various iron sources when the iron was added directly to the test diet and when the iron was a component of a food which was then added to the test diet. Foods were processed in several ways. For example, farina enriched with fe r r i c 16 orthophosphate was cooked and ferrous sulfate was added to a biscuit mix and baked into biscuits. It was concluded that processing had l i t t l e or no effect on the bioavailability of iron. Pla et a l . (1973) used a hemoglobin repletion assay with chicks and rats to determine the RBV of iron from bread. Bread was baked from flour enriched with ferrous sulfate and reduced iron. It was found that there was no evidence for any effect of baking on the bioavailability of the iron sources. Coccodrilli et a l . (1976) determined the RBV of elemental iron powders which were commercially processed into a bran flake breakfast cereal. It was observed that commercial cereal processing procedures did not significantly alter the expected RBV of the iron sources to rats. Verma et a l . (1977) baked bread with flour containing different forms of elemental iron. It was found that the iron sources had similar availability to rats whether they were added directly to the diet or baked into bread and then added to the diet. In a recent study, Schricker and Miller (1982) used an in vitro method to estimate the relative a v a i l a b i l i t y of various iron f o r t i f i c a t i o n sources. The effect of baking on f o r t i f i c a t i o n iron a v a i l a b i l i t y in bread was evaluated. The different forms of iron were added to the bread either before baking or during homogenization of the baked bread. The bread was analyzed alone or as part of a complex meal. A reduction in available iron was observed as a result of baking with the complex meal but not with the bread meal. The literature indicates that processing generally increased the bioavailability of added iron when the process involved heating a predominately aqueous food (Clemens and Mercurio, I98I; Wood et a l . , 17 1978; Disler et a l . , 1975; Theurer.et a l . , 1971. 1973). Dry heat processing was found to have l i t t l e effect on the bioavailability of added iron (Verma et a l . , 1977; Coccodrilli et a l . , 1976; Pla et a l . f 1973; Fr i t z and Pla, 1972). There is l i t t l e information i n the literature on the effect of heat processing on the bioavailability of iron naturally present in foods. Sayers et a l . (1973) measured the absorption of intr i n s i c and added iron in three different vegetable foods using a double radioiron method with iron deficient human subjects. Maize was prepared as porridge, soya beans as biscuits and wheat as whole wheat bread. A l l of the foods were combined with ascorbic acid so that differences in bioavailability due to baking per se could not be determined. It was found that iron in cooked maize porridge increased in bioavailability when cooked in combination with ascorbic acid. However, iron in baked soya biscuits or baked bread did not change in bioavailability when cooked with ascorbic acid. 10. EFFECT OF BRAN PARTICLE SIZE ON IRON AVAILABILITY A review of the literature indicates that the effect of bran particle size on the ava i l a b i l i t y of iron from bran has not been considered. It is hypothesized by this author that a decrease in the particle size of bran may increase the amount of iron available for absorption by increasing the surface area of the fiber exposed to the digestive process. 18 11. EFFECT OF ORANGE JUICE ON THE AVAILABILITY OF IRON . The a b i l i t y of orange juice to enhance iron a v a i l a b i l i t y has been well documented (Monsen, 1982; Schricker and Miller, 1982; Rossander et a l . , 1979). Orange juice contains a variety of organic acids including c i t r i c , ascorbic, malic and oxalic. C i t r i c acid is the most predominant organic acid in orange juice. The average content is 0 .60 to 0 .75 S of c i t r i c acid per 100 ml (Nagy, 1978; Clements, 1964) . The content of ascorbic acid in fresh orange juice i s about 5 ° mg per 100 ml of juice (Health and Welfare Canada, 1979) . The a b i l i t y of ascorbic acid to enhance nonheme iron absorption from foods has been extensively studied. Monsen (1982) recently reviewed the roles of orange juice and ascorbic acid in nonheme iron absorption. It is interesting that most studies which have demonstrated the a b i l i t y of orange juice to enhance nonheme iron a v a i l a b i l i t y have attributed the absorption promoting property of the orange juice to the presence of ascorbic acid and have not considered the possible influence of c i t r i c acid (Monsen, 1982; Rossander et a l . , 1979)-However, a recent study by Kojima et a l . (I98I) examined the effects of ascorbic acid, c i t r i c acid and orange juice on the solubilization of iron from cooked pinto beans using an in vitro method. In one experiment, a dual incubation procedure was carried out at pH 2 and then pH 6 . It was found that a 10 mM solution of ascorbic acid solubilized h% of the iron from a cooked pinto bean suspension. Virtually a l l of the soluble iron was present in the ferrous state. A 10 mM solution of c i t r i c acid solubilized about 50% of the iron and 2% of the soluble iron was in the ferrous form. It was found that the effects of ascorbic acid and 19 c i t r i c acid were additive and 10 mM solutions of each combined solubilized approximately 70% of the iron from a cooked pinto bean suspension. Almost a l l of the soluble iron was in the ferrous state. The effect of orange juice was also investigated using a dual incubation procedure. The solubilization of iron from a cooked pinto, bean suspension in the presence of fresh orange juice was as great as a combination of 10 mM c i t r i c acid and 10 mM ascorbic acid. Kojima et a l . (1981) concluded that while other factors may play a role, the solubilization of iron from beans by orange juice is probably due to the effects of c i t r i c and ascorbic acids. A survey of the literature indicates that few studies have looked at the effect of orange juice and i t s constituent organic acids on the a v a i l a b i l i t y of iron from wheat or wheat bran. Kuhn et a l . (I968) examined the effect of chelates on food iron absorption using a double radioiron tag method on humans. Wheat which had been i n t r i n s i c a l l y labelled was cooked and fed as a small cake. Two moles of ascorbic acid per mole of iron were added immediately before oral administration. Comparison of the absorption of iron from wheat in the same individual with and without ascorbic acid showed that ascorbic acid increased absorption 2.5 times. Sayers et a l . (1973) determined the absorption of iron from maize, wheat and soya bean using a double radioiron method in iron deficient human subjects. The addition of 50 mg ascorbic acid to unsupplemented maize porridge increased the average iron absorption from about 7% to Zk/o. The addition of ascorbic acid to soya biscuits or whole wheat bread prior to baking did not affect the absorption of i n t r i n s i c or extrinsic iron. It was suggested that the lack of effect was probably 20 due to the fact that ascorbic acid was destroyed during baking. 12. METHODS FOR THE DETERMINATION OF IRON AVAILABILITY Many methods exist for determining the quantity of iron available for absorption from foods and these have been reviewed in several textbooks (Beutler, 1980; Hallberg, I98O; Narins, I98O). The oldest method for measuring food iron absorption is the chemical balance technique. By this method absorption equals the difference between oral intake and fecal loss. Generally, the technique is tedious, insensitive and inaccurate. The very small amount of iron absorbed relative to the iron content of the diet makes the difference d i f f i c u l t to measure with precision. As well, differentiation between excreted and unabsorbed iron is not possible (Morck and Cook, I 9 8 I ; Beutler, I 98O; Narins, I 9 8 O ) . The most frequently used animal bioassay is the rat hemoglobin repletion test. The method is described by F r i t z et a l . (1978, 1974) and i t has been adopted as the standard method of analysis for the bioavailability of iron by the Association of O f f i c i a l Analytical Chemists (AOAC, 1980, sections 43.217 - 43.219). In this test, rats are made anemic on a low iron basal diet. The anemic animals are then fed a diet containing graded levels of iron provided by either a reference standard of highly available ferrous sulfate or the test source. After two weeks the hemoglobin response to the test source of iron is compared to the response observed in a comparable group of animals fed identical levels of iron from the reference standard. The results are expressed as the relative biological value (RBV) of the test iron source relative to ferrous sulfate which is given a value of 100. 21 There are advantages and disadvantages to the use of animal assays such as the hemoglobin repletion test. The principal advantage is the use of an intact biological system (Miller and Schricker, 1982). However, there are several disadvantages which include the problems associated with extrapolating results from rats to humans. Fritz and Pla (1972) reported that the absorption enhancing effect of ascorbic acid has not been demonstrated in rats. They postulated that species such as rats, which do not need a dietary source of ascorbic acid, produce enough ascorbic acid in their bodies to minimize any effect from dietary sources. It has also been reported that while humans preferentially absorb ferrous iron; f e r r i c and ferrous iron are equally well absorbed by rats (Narins, 1980). As well, the hemoglobin repletion test probably overestimates the RBV of iron sources because anemic animals use iron more eff i c i e n t l y . However, i t is necessary to use depleted animals in order to measure a response to dietary iron (Miller and Schricker, 1982). The requirement for graded levels of iron in the diets means that when whole foods are used the composition of diets i s usually not constant between groups (Miller and Schricker, 1982). Finally, animal absorption studies are expensive. Radioisotope techniques allow the most precise measurements of dietary iron absorption in humans (Narins, I98O). Isotopic techniques are also used with animals but the disadvantages of animal studies have already been discussed. Radioisotopes were f i r s t used to measure the absorption of iron from single foods which had been i n t r i n s i c a l l y labelled with radioactive iron prior to harvest or slaughter. Studies with i n t r i n s i c a l l y labelled 22 foods demonstrated that iron absorption from different foods varied widely and that absorption was greater from foods of animal origin (Hallberg, 1980; Narins, 1980). However, intrinsic studies provided l i t t l e information about iron absorption from a complete meal. A major advance in the study of the bioavailability of iron was made when the extrinsic tag method was demonstrated to be a valid technique (Hallberg, I98O). It has been determined that a l l of the nonheme iron of foods ingested in the same meal becomes part of a common nonheme iron pool in the gut. Absorption of nonheme iron can be measured by adding a radioactive inorganic iron tracer to a food or composite meal. Isotopic exchange between the tracer and the native nonheme iron results in complete and uniform labelling of the nonheme iron (Hallberg, I98O). Whole body retention of radioiron is the most reliable, sensitive and quantitative method for measuring iron absorption in humans. The general procedure is to count the natural background radioactivity in the* fasting subject. The subject then ingests food containing a known amount of radioiron. The radioactivity of the whole body is counted after 4 hours and after 14 to 20 days when the unabsorbed iron has been eliminated from the body (Narins, I98O). Radiolabelled heme iron exchanges completely with the pool of heme iron in the gut. Therefore, i t is possible by using the two pool extrinsic radioiron tag method to measure the absorption of heme and nonheme iron in the same meal (Beutler, 1980; Hallberg, 1980). Human studies using radioiron have proven to be very successful. However, such studies are expensive, complicated and time consuming. As well, the administration of radioisotopes to humans i s not always possible or desirable (Miller and Schricker, 1982). 23 13. IN VITRO METHODS FOR THE DETERMINATION OF IRON AVAILABILITY In vitro methods offer an appealing alternative to human and animal studies and they have been used to estimate iron a v a i l a b i l i t y for at least 50 years (Miller and Schricker, 1982). In vitro methods can be simple, rapid and inexpensive (Miller et a l . , I98I). While human and animal studies are necessary to measure actual absorption and u t i l i z a t i o n of food iron, in vitro techniques can be used to measure the amount of dietary iron that i s potentially available for absorption (Narasinga - Rao and Prabhavathi, 1978). The determination of ionizable iron has been used as an estimate of available iron. Ionizable iron has been determined as that fraction of the total iron in a food that w i l l react with a complexing agent such t as< < , o t -dipyridyl, t r i p y r i d y l triazine or bathophenanthroline to form a chromagen which can be quantified spectrophotometrically (Miller and Schricker, 1982; Narins, I98O). Schackleton and McCance (1936) estimated the ionizable iron of a variety of foodstuffs by extraction with a sodium acetate-acetic acid buffer at pH 5.5 in the presence of sodium hydrosulfite, followed by measurement of the coloured iron complex formed with U, -dipyridyl. Leichter and Joslyn (I967) used a similar procedure to estimate available iron in 3 different types of bread. However, the determination of ionizable iron in foods is probably of l i t t l e value since the conditions in the gastrointestinal tract which influence iron a v a i l a b i l i t y are not considered (Narasinga-Rao and Prabhavathi, 1978). 24 Other in vitro methods of predicting the av a i l a b i l i t y of dietary iron have concentrated on duplicating the chemical environment of the stomach with subsequent measurement of the soluble iron released by simulated gastric digestion. Jacobs and Greenman (1969) estimated the amount of iron released from a number of foods by peptic digestion. Various foods were incubated with pepsin and hydrochloric acid. The digest was centrifuged and the supernatant analyzed for ionizable iron by a tri p y r i d y l triazine method. Total soluble iron, which measured both ionizable and complexed iron, was also determined. It was considered that the quantities found represented the amount of iron expected to be present in soluble form in the stomach and therefore available for absorption. Hart (1971) used an analysis similar to that employed by Jacobs and Greenman (1969) to compare the a v a i l a b i l i t y of iron in enriched white bread with that of iron naturally present in wholemeal bread. In another in vitro method, Lock and Bender (I98O) measured the proportion of iron liberated from 20 foods after incubation with human gastric juice. The digest was centrifuged and the supernatant analyzed for soluble iron by atomic absorption spectroscopy. It was suggested that the iron solubilized by human gastric juice was "chemically available". However, the use of simulated gastric digestion probably does not determine the true av a i l a b i l i t y of iron since most dietary iron is absorbed from the small intestine and not from the stomach (Narasinga-Rao and Prabhavathi, 1978). Narasinga-Rao and Frabhavathi (1978) developed an in vitro method for the determination of available nonheme iron in foods. Their method attempted to simulate the chemical environment of the stomach and small 25 intestine. Food was subjected to treatment with pepsin and hydrochloric acid at pH 1 .35 at 37°C for 1.5 h. The solution was subsequently adjusted to pH 7 ' 5 with sodium hydroxide and f i l t e r e d . Ionizable iron was determined in the f i l t r a t e by reaction with '-dipyridyl. Narasinga-Rao and Prabhavathi (1978) used their method to estimate the a v a i l a b i l i t y of iron from various foods and diets. It was reported that the percent ionizable iron at pH 7>5 in a number of diets correlated highly with percent iron absorption from the same diets observed in adult males. It was also observed that the ionizable iron at pH 7 ' 5 increased in the presence of ascorbic acid and meat extract while i t decreased in the presence of phytate and tannins; similar to the effects of these factors on iron absorption in human subjects. Hazell et a l . (1978) used an in vitro digestion procedure, which simulated the physiological conditions of the stomach and the small intestine, to investigate the a v a i l a b i l i t y of iron from meat. Another in vitro method to estimate food iron a v a i l a b i l i t y was developed by Miller et a l . (I98I). They reported that their method differs from previously published in vitro methods such as that of Narasinga-Rao and Prabhavathi (1978) in two important ways. The pH adjustment from gastric to intestinal levels is gradual and reproducible and only low molecular weight soluble iron i s used in the estimation of available iron. Recognition of these important differences in methodology prompted the selection of the in vitro method of Miller et a l . (1981) for use in this study. Details of the method have been described elsewhere (Miller and Schricker, 1982; Miller et a l . , 1981; Schricker et a l . , I 9 8 I ) but for the purpose of discussion the method, as adapted for this study, w i l l be outlined b r i e f l y . 26 The method involved digesting aliquots of homogenized meals, adjusted to pH 2 and mixed with pepsin in a 37 G shaking water bath for 2 h. After the pepsin digestion, aliquots were removed for measurement of titratable acidity. The pH of the pepsin digests was then raised to 7 using sodium bicarbonate. An amount of sodium bicarbonate equivalent to the titratable acidity (the amount of potassium hydroxide required to raise the pH of pepsin digest to 7*5) was placed in a segment of dialysis tubing (mol. wt. cutoff 6000 to 8000) which was placed in the digest. A second 2.5 h incubation at 37 C in a shaking water bath was conducted in the presence of a pancreatin-bile extract mixture. At the end of the incubation, the dialysis bags were removed, rinsed and the dialysate weighed and analyzed. Dialysable iron was determined by a modified bathophenanthroline method (Clark, 1962; Lee and Stumm, I960; Smith et a l . , 1952). Total iron was determined in the muffin samples by atomic absorption spectroscopy according to the method of Maurer (1977). In the original method, dialyzable iron was also determined by counting ^ Fe activity. Miller et a l . (1981) reported that the two methods for the determination of dialyzable iron were highly correlated. In the original method, food samples were analyzed for both heme and nonheme iron. However, Miller (1982) noted that in meals which do not contain meat, the amount of heme iron i s insignificant and nonheme and total iron are therefore identical. Miller et a l . (1981) discussed their rationale for choosing various experimental procedures. They explained that the use of a bicarbonate solution in dialysis tubing allows for a slow increase in pH before and during pancreatin digestion. As well, this method of pH adjustment permits dialysis of iron to occur during the neutralization process and thus 27 more closely parallels the in vivo situation. The use of dialysis tubing of a specific pore size permits discrimination between high and low molecular weight soluble iron complexes and i s based on the assumption that absorbable iron i s present in the duodenum as a low molecular weight soluble chelate. Schricker et a l . (1981) compared the in vitro method of Miller et a l . (I98I) with rat and human extrinsic tag methods for estimating food iron a v a i l a b i l i t y . Complex meals were used in the comparison. When the criterion for comparison was the a b i l i t y to show s t a t i s t i c a l l y significant differences between iron a v a i l a b i l i t y in the various meals, there was substantial agreement between the in vitro and human in vivo methods. An in vitro method, such as that developed by Miller et a l . (I98I), has certain limitations. These include uncertainties over the use of an a r t i f i c i a l system, the inexact duplication of in vivo conditions and the ina b i l i t y to account for the effects of such factors as active transport and brush border binding proteins (Miller and Schricker, 1982). As a result of these limitations, Miller et a l . (1981) noted that in vitro measurement of iron a v a i l a b i l i t y must be a relative rather than an absolute indication of av a i l a b i l i t y . However, an in vitro method for the estimation of relative iron a v a i l a b i l i t y i s rapid and inexpensive. Other advantages of a method such as that developed by Miller et a l . (1981) include reduced var i a b i l i t y compared to in vivo methods and the a b i l i t y to precisely control conditions during the determinations (Miller and Schricker, 1982). 28 lk. CONSUMPTION OF WHOLE GRAIN PRODUCTS The consumption of foods made with unrefined cereals and their •byproducts, such as bran, is increasing (Turnlund, 1982). This may be illustrated by the number of "new" whole grain breakfast cereals that have recently been introduced into supermarkets. The Health Protection Branch of Health and Welfare Canada has recommended that more whole grain foods be included in the Canadian diet (Health and Welfare Canada, 1977)' Recommendations for the consumption of whole grain products have been made for several reasons. Whole grain products contain not only calories and other nutrients commonly associated with cereals, but also significant amounts of essential minerals which may not be present in refined cereal products (Turnland, 1982; O'Dell et a l . , 1972). Kent (1975) reported the percentage of the total constituents of wheat present in the main morphological parts (Table I i ) . It was found that 67, 23 and 10% of the total mineral content was present in the bran, endosperm and germ respectively. O'Dell et a l . (1972) analyzed the distribution of elements among the morphological components of the wheat kernel and determined that approximately 15 «1» 27.6, 78.6 and of the total iron was present in the germ, endosperm, aleurone layer and h u l l respectively. The outer layers of wheat consisting of the bran (composed of the hull and aleurone layer) and the germ with their high content of minerals are removed during milling (Turnlund, 1982). The content of several essential trace elements in wheat, wheat fractions and wheat flour is compared in Table III. Only some of the trace nutrients are added back during f o r t i f i c a t i o n . In Canada, wheat flour i s f o r t i f i e d with thiamine, 29 Table II. Percentage of the total constituents of wheat present in the main morphological parts 3 . Part Weight Constituents (g/100 g grain) Starch Protein Fiber Fat Minerals Bran 15 0 20 93 30 67 Endosperm 82 100 72 k 50 23 Germ 3 0 8 3 20 10 aAdapted from Kent (1975). 30 Table III. The mineral content of wheat and milled fractions ppm Fe Zn Mn Gu Wheat 18-31 21-63 24-37 1 . 8 - 6 . 2 Bran 74-103 56-141 72-144 8 . 4 - 1 6 . 2 Germ 41-58 100-144 101-129 7 .2-11.8 Flour 3 . 5 - 9 . 1 3 . 4 - 1 0 . 5 2 . 1 - 3 . 5 O.62-O.63 aAdapted from Turnlund (1982) 31 riboflavin, niacin and iron (Health and Welfare Canada, 1981). Interest in unrefined cereals and their byproducts has also increased because numerous hypotheses have linked increased consumption of whole grains and dietary fiber to beneficial health effects. Several authors have reviewed the effects of dietary fiber intake on man (Kelsay, 1978; Burkitt, 1977; S p i l l e r and Amen, 1974). A low fiber intake has been reported to be associated with ischemic heart disease, diabetes, constipation, appendicitis, hiatus hernia, hemorrhoids, diverticular disease of the colon and cancer of the colon (Kelsay, 1978; Burkitt, 1977; S p i l l e r and Amen, 1974). The claims of health benefits have resulted in widespread recommendations to include more fiber in the diet (Turnlund, 1982). The addition of wheat bran to such products as bread, cakes, muffins and cookies i s one method of increasing the dietary fiber content of the diet (Polizzoto et a l . , 1983). Dietary fiber is a loosely defined group of polymers present in varying degrees in a l l plants and natural plant products (i.e., grains, vegetables, and f r u i t s ) . The most important of these polymers are cellulose, hemicellulose, lignin and pectin. Cellulose is a high molecular weight polymer of repeating units of/6-linked glucose. Hemicelluloses are made up of long chains of such monosaccharides as xylose, galactose, glucose, mannose and arabinose. Pectic substances are mostly polymers of galacturonic acid. Lignin i s made up of repeating phenyl-propane units and is not a carbohydrate (Spiller and Amen, 1974). Various methods exist for the determination of dietary fiber in foods including the determination of crude fiber, acid detergent fiber (ADF) and neutral detergent fiber (NDF). These and other methods have been discussed 32 by Kelsay (1978). Anderson and Clydesdale (1980a) analyzed the total dietary fiber content of a standard wheat bran using various methods. The results of their analyses are shown in Table TV. Anderson and Clydesdale (1980a) also reported that the dietary fiber of a standard wheat bran was composed of 28.3, 8.7, 3'2 and hemicellulose, cellulose, l i g n i n and pectin respectively. Although wheat bran i s being consumed in the diet, very l i t t l e i s known about the a v a i l a b i l i t y of the endogenous iron of this fiber source. 15. AVAILABILITY OF THE ENDOGENOUS IRON OF WHEAT BRAN Wheat bran contains relatively high levels of iron when compared with other foods as shown i n Table V. However, a survey of the literature reveals that l i t t l e work has been done on the avai l a b i l i t y of iron from bran. Some studies have compared the absorption of iron from whole grain and refined wheat products (Ranhotra et a l . , 1979; Bjorn-Rasmussen, 1974). Since the iron in whole grain flour i s provided by both bran and endosperm, the av a i l a b i l i t y of iron from bran i t s e l f cannot be ascertained. For example, Bjorn-Rasmussen (1974) measured the percentage iron absorption from r o l l s with and without added bran. Turnlund (1982) calculated the results to reflect the amount of iron absorbed in mg (Table VI). These results showed that the total amount of iron absorbed remained essentially the same while the percentage bran added to the r o l l s increased. However, because bran contains a significant amount of iron, the percentage absorption of iron decreased. 33 Table IV. Composition of total dietary fiber of a standard wheat bran by various methods Method Dietary fiber content (fo) Fractionation 44.12 Enzymatic 46.0 Neutral detergent fiber 40.2 Acid detergent fiber 11.9 Crude fiber 8.91 aAdapted from Anderson and Clydesdale (1980a). 3k Table V. Iron content of some foods calculated from tabular values' Food mg iron/ 100 g edible portion Milk, whole f l u i d trace Brown ri c e , cooked 0.5 Cheddar cheese 0.7 Tuna, canned 1-9 Peanut butter 1.9 Whole egg 2.2 Pork, roasted 2.9 Ground beef, cooked 3.2 Raisins 3.6 Liver, calf, cooked Ik A Wheat bran 15.0 Molasses, blackstrap 16.0 h e a l t h and Welfare Canada, 1979* 35 Table VI. Iron absorption by human subjects from r o l l s Percent bran added to r o l l s Iron content of flour and bran (mg/lOOg) Absorption % mg 0 . 0 0.48 17 .8 0.085 1-7 0 .62 14.2 0.088 3-3 0.75 14.3 0.107 6 .7 0 .93 6 . 3 0.059 1 0 . 0 1.20 8 . 5 0.102 aAdapted from Turnlund (1982) based on data reported by Bjorn-Rasmussen (1974). 36 Many other studies have looked at the effect of adding wheat bran to a meal on the absorption of dietary nonheme iron (Reinhold, 1982; Reinhold et a l . , 1981; Simpson et a l . , 1981). Studies such as these have provided l i t t l e information about the ava i l a b i l i t y of iron from bran per se. However, i t i s generally agreed that the ava i l a b i l i t y of iron from whole grains and wheat bran is poor (Turnlund, 1982; Erdman, I98I). Hussain et a l . (I965) examined the avail a b i l i t y of iron from in t r i n s i c a l l y labelled wheat. Whole wheat flour was cooked into pancakes and fed to human subjects. The average absorption of wheat iron was h.% in 21 healthy subjects and 7*8% in 21 iron deficient subjects. Elwood et a l . (1968) determined the absorption from bread by humans of the iron naturally present in bran. Bread was made from white flour with added i n t r i n s i c a l l y labelled bran. It was found that iron from wheat bran was less well absorbed than iron from f e r r i c ammonium citrate added to bread made from white flour. In another study, Layrisse and Martinez-Torres (1971) examined the absorption of iron from various i n t r i n s i c a l l y labelled foods and found that the absorption from wheat was 5%« Finally, Erdman (1981) reported that various components of bran including phytate, phosphate and dietary fiber have the a b i l i t y to chelate minerals. It was suggested that the availability of iron from bran may depend on the digestability of the various iron complexes formed with the bran components. 37 III. MATERIALS AND METHODS 1. BRAN Hard red wheat bran was obtained from the American Association of Cereal Chemists (AACC, St. Paul, MN). Upon receipt of the bran some was sifted to obtain two different particle sizes. Five hundred g of bran was sifted on a series of Nalgene sieves f i t t e d with a pan and cover, for 15 min using an automatic shaker (Eberbach Corporation, Ann Arbor, Ml). The bran which passed through mesh No. 10 (2.00 mm diameter) but which was retained on mesh No. 20 (1.00 mm diameter) was collected and subsequently referred to as #10 bran. After sufficient #10 bran was obtained some was ground in a porcelain b a l l m i l l to further decrease the particle size. After grinding, the bran was sifted and that portion which passed through mesh No. 40 (0.0425 mm diameter) was collected and subsequently referred to as #40 bran. The bran was very d i f f i c u l t to grind and i t took 72 to 96 h to obtain enough #40 bran for one muffin batch. This limited the number of experiments that could be done with #40 bran. It was assumed that the porcelain b a l l m i l l did not contribute iron to the bran. Samples of bran were analyzed for moisture content and total iron. The bran was stored in plastic bags at approximately 4 C. The particle size distribution of each bran fraction (#10 and #40 bran) was determined by a method similar to that recently described by Polizotto et a l . (1983)-38 2. MUFFINS A l l of the muffins used for this study were prepared by combining the ingredients lis t e d in Table VII. The muffins were made using either one of the two particle sizes of bran. The muffin product was formulated so as to eliminate, as much as possible, a l l sources of dietary iron except that contributed by the wheat bran. The chemical composition of the muffin dough is shown in Table VIII. The moisture content and total iron of the unenriched flour were verified experimentally. For each experiment, two separate batches of muffin dough were prepared. For each muffin batch, half of the dough was baked and half was unbaked. Paper-lined muffin tins were f i l l e d with about 65 g portions of dough. The muffins were baked in a preheated oven (Despatch Oven Company, Minneapolis, MN; at approximately 200 C for 20 min. After cooling at room temperature for 45 min the muffins were frozen i n plastic bags at about -25*C. The unbaked dough was divided into approximately 65 g portions and immediately frozen in plastic bags. Samples of muffin dough and cooled, baked muffin were taken for moisture determination prior to freezing. 3. MOISTURE DETERMINATION The moisture content of bran and flour was estimated by drying o a known weight of the sample in a vacuum oven at 100 C for approximately 5 hours as described by A0AC (A0AC, 1980, sections 14.002 - 14.003). The analyses were in t r i p l i c a t e and mean values were calculated. The moisture content of baked and unbaked muffins was estimated by mixing a known weight of sample with quartz sand and water and heating on a hot plate u n t i l almost dry as described by AACC (AACC, 1978, section 39 Table VII. Muffin recipe 235 g f l u i d skim milk cL 230 g unenriched , untreated wheat flour 150 g AAGG Certified hard red wheat bran 75 g corn o i l 60 g granulated white sugar 0 50 g whole egg 16 g baking powder1 2 g s a l t 6 aPurity brand donated by Maple Leaf M i l l s , Toronto, Ont. ^Townhouse brand, Empress Foods Ltd., Vancouver, B.C. BC Sugar, Vancouver, B.C. Magic brand, Standard Brands Canada Ltd. eWindsor brand, Canadian Salt Co., Ltd., Montreal, Que. 4o Table VIII. Chemical composition of muffin dough based on tabular i a,b values Ingredient Weight (g) Moisture Energy Protein (%) (kcal) (g) Fat (g) Total CHO (g) Iron (mg) f l u i d skim milk 235 90 82 8.2 t r 11.8 0.091 unenriched flour 230 12 837 24.2 2.3 175.0 1.8 wheat bran^ 150 9 254 24.9 2.0 99.0 24.75 corn o i l 75 0 663 0.0 75.0 0.0 0.0 white sugar 60 t r 231 0.0 0.0 59.7 0.057 whole egg 50 75 79 6.0 6.0 1.0 1.1 c e baking powder ' 16 t r 21 t r t r 5.0 -salt 2 0 0 0.0 0.0 0,0 0.002 Total 818 186 2167 63.3 85.3 351.5 27.80 h e a l t h and Welfare Canada (1979). ^0 * none, t r = trace, (-) = no suitable value found but measurable amount may be present. CUSDA (1975)-dAACC analysis. Contains starch, monocalcium phosphate and sodium bicarbonate. 41 o 44-60). The samples were then dried in a vacuum oven at 100 G for approximately 5 h* Duplicate determinations for both baked and unbaked muffin samples were made for each muffin batch. 4. IN VITRO ESTIMATION OF RELATIVE IRON AVAILABILITY Since adaptations were made to the method of Miller et a l . (1981) to accommodate the conditions and f a c i l i t i e s existing in the Department of Food Science at the University of B r i t i s h Columbia and because the in vitro determination i s new and interesting, the protocol followed by this author w i l l be described in d e t a i l . A flow diagram of the method is shown in Figure 1. 4.1 Reagents and Materials Water - The water used throughout the analyses was d i s t i l l e d , deionized (d.d.) water. Glassware - A l l glassware was washed, rinsed withd.d. water, soaked overnight in IN HCL and rinsed again with d.d. water. Orange juice - Bel-Air brand frozen orange juice concentrate (Empress Foods Ltd., Vancouver, B.C.). The orange juice was reconstituted with three volumes of d.d. water. Ascorbic acid - (Matheson, Coleman and B e l l Manufacturing Chemists, Norwood, OH). Ci t r i c acid - (Matheson, Coleman and B e l l Manufacturing Chemists, Norwood, OH). kz Figure 1. Schematic of the in vitro method for determination of dialyzable iron. Sample blended to slurry i I Homogenate adjusted to pH 2 with 6N HCL Incubation with pepsin 0 2h, 37 C shaking water bath I J i f I Titratable j Acidity j I I Incubation with pancreatin-bile 0 30 min, 37 G shaking water bath 1 i ^ — A d d i t i o n of dialysis bag V Incubation with pancreatin-bile 0 2 h, 37 C shaking water bath I Collection of dialysate pH 7-7*5 Analysis of dialysate % Dialyzable iron k3 Iron reference solution - Certified atomic absorption standard (Fisher Scientific Co., Fair Lawn, NJ). Pepsin - The solution was prepared just before use. Sixteen g pepsin (from porcine stomach mucosa, sigma P700, Sigma Chemical Co., St. Louis., MO) were suspended in 0.1N HCL and brought to 100 ml with 0.1N HCL. Pancreatin-Bile - The solution was prepared just before use. Four hundred mg pancreatin (from porcine pancreas, Sigma P1750, Sigma Chemical Co., St. Louis, MO) and 2.5 g b i l e extract (porcine, Sigma B8631, Sigma Chemical Co., St. Louis, MO) were dispersed in 0.1 M sodium bicarbonate (Fisher Scient i f i c Co., Fair Lawn, NJ) and brought to 100 ml with 0.1 M sodium bicarbonate. Bathophenanthroline solution - A 0.001 M solution was prepared by dissolving 0.0334 g of bathophenanthroline (4,7-diphenyl-l,10-phenanthroline) (Sigma B1250, Sigma Chemical Co., St. Louis, M0) in 50 ml of ethyl alcohol. The solution was heated in a water bath to dissolve the reagent, cooled and brought to 100 ml with d.d water. Hydroxylamine Hydrochloride - Reagent grade hydroxylamine hydrochloride contains appreciable amounts of iron. An adaptation of the method described by Smith et a l . (1952) was used to purify the hydroxylamine hydrochloride. One hundred ml of a 20% aqueous solution of hydroxylamine hydrochloride (Sigma H9&76, Sigma Chemical Co., St. Louis, M0) were prepared and placed in a 250 ml conical separatory funnel. Three to 4 ml of 0.001 M bathophenanthroline were then added. Next, 10 to 20 ml of chloroform (BDH Chemicals, Toronto) were added to the separatory funnel, the contents of the funnel were shaken and the layers allowed to separate for 5 min. Then the colored bottom layer containing the 44 chloroform and extracted red iron impurity was drawn off and discarded. The separation was repeated u n t i l a colorless solution of hydroxylamine hydrochloride was obtained. The iron-free hydroxylamine hydrochloride solution was stored in a glass bottle at 4°C. Protein precipitant solution - The solution was prepared just before use. Ten g trichloroacetic acid (Fisher A322, Fisher Scient i f i c Co., Fair Lawn, NJ), 50 ml 20% purified hydroxylamine hydrochloride and 10 ml concentrated HCL were combined and brought to 100 ml wilh d.d water. Chromagen solution - The solution was prepared just before use. Twenty -five mg bathophenanthroline sulfonate (4,7-diphenyl-l,10-phenanthroline disulfonate) (Sigma 13?5» Sigma Chemical Co., St. Louis, M0) were dissolved in 4M sodium acetate in a 100 ml volumetric flask. The solution was o heated to approximately 80 C in a water bath to dissolve the reagent, cooled and brought to 100 ml with 4M sodium acetate. Dialysis tubing - Spectrapor brand dialysis tubing with molecular weight cutoff of 6000-8000 (Fisher 08-670C, Fisher Scientific Co., Fair Lawn, NJ) was cut into 15 cm lengths and soaked in d.d. water u n t i l required. Sample bottles - Wheaton brand glass, 120 ml bottles with plastic snap caps (Fisher 03-335 10D, Fisher Scient i f i c Co., Fair Lawn, NJ) were used for the pancreatin-bile digestion. Atomic absorption spectrophotometer - A Perkin Elmer model 4000 (Perkin Elmer, Norwalk, CT) double beam atomic absorption spectrophotometer with air-acetylene flame was used to measure iron at 248.3 nm. ^5 4.2 Procedure The general plan of the experimental procedure i s shown in Figure 2. For each muffin batch the in vitro analysis was carried out separately at the same time on baked and unbaked samples. One slurry was prepared using baked muffins and a separate slurry was made with unbaked muffin dough. 4.2.1 Slurry preparation Frozen muffin samples were thawed at room temperature prior to blending. The slurry components were homogenized in a Waring blender (Waring Products Division, Dynamics Corp. of America, New Hartford, CT) to a creamy consistency. The slurry was adjusted to pH 2 with 6N HCL while s t i l l in the glass blender jar. With the blender running at a low speed, acid was added to the slurry dropwise by pipette. At intervals, the blender was stopped and the pH of the slurry determined using an Accumet brand pH meter (Fisher Scient i f i c Co., Fair Lawn, NJ). 4.2.2 Pepsin digestion From each slurry, 2x lOOg aliquots of muffin mixture were transferred to 250 ml erlenmeyer flasks. Ten ml of the pepsin suspension were added to each flask (Since Sigma P700 pepsin i s about 32°% pepsin, this is equivalent to about 0.5 g pepsin). The flasks were sealed with parafilm o and aluminum f o i l and the slurries were incubated for 2 h in a 37 C shaking water bath (Magni-Whirl, Blue M Electric CO., Blue Island, LL). After incubation, samples were removed from each of the flasks of pepsin digest as follows: l ) Twenty g aliquots of the pepsin digest were transferred to each of 3*120 ml snap cap bottles. These aliquots 46 Figure 2. General plan of the experimental procedure Experiment 1 Muffin Batch A Muffin Batch B continues as for Muffin Batch A Baked Muffins Unbaked Muffins A continues as for unbaked muffins Dialysate 2 47 were frozen u n t i l required for the pancreatin-bile digestion. 2) A fourth 20 g aliquot was transferred to a 30 ml beaker. Each beaker was sealed with parafilm and aluminum f o i l and the samples were frozen u n t i l required for the determination of titratable acidity. 4 . 2 . 3 Titratable acidity c The frozen aliquots were thawed in a 37 C water bath. Five ml of the pancreatin-bile suspension were added to each beaker of thawed digest. Using a pH meter, the titratable acidity was determined by t i t r a t i n g the combined pepsin digest pancreatin-bile extract mixture with 0.5N potassium hydroxide (Fisher S c i e n t i f i c Co., Fair Lawn, NJ) to pH 7-5 and recording the volume of potassium hydroxide required. 4 . 2 . 4 Preparation of dialysis bags Each prepared dialysis bag contained a total volume of 20 ml. The volume of each dialysis bag was made up of an amount of 0.5N sodium bicarbonate (Fisher S c i e n t i f i c Co., Fair Lawn, NJ) equivalent to the amount of potassium hydroxide used in the titratable acidity determination and d.d. water. The dialysis bags were stored in a sodium bicarbonate solution of equivalent concentration u n t i l just before use. String was used to close the bags. 4 . 2 . 5 Pancreatin-bile digestion The frozen aliquots were thawed in a 37 C shaking water bath. Then one prepared dialysis bag was placed in each bottle of thawed digest and the bottles were sealed with plastic caps. Care was taken that each bag was completely covered with pepsin digest. The samples were incubated 48 in a 37 "c shaking water bath for 30 min. Then, 5 nil of the pancreatin-bile solution were added to each bottle of digest and the incubation was continued for an additional 2 h. 4.2.6 Collection of dialysate At the end of the incubation period, the dialysis bags were removed from the digest and rinsed by dipping in d.d. water. One end was cut off each bag and the contents were transferred into a preweighed test tube. The test tube was then reweighed and the weight of the dialysate determined by difference. 4.2.7 Analysis of dialysate Bathophenanthroline reactive iron was measured in the dialysates immediately after the pancreatin-bile incubation. A 10 ml aliquot of each dialysate was transferred to a clean test tube. Five ml of the protein precipitant solution were added to each tube of dialysate. Then the tubes were covered with parafilm and aluminum f o i l and were o l e f t to s i t overnight at approximately 4 C. The next morning the tubes of dialysate were removed from refrigeration. The addition of the protein precipitant solution had caused the samples to turn cloudy. To obtain a clear solution for analysis, the tubes were heated in a boiling water bath for 10 min. After cooling to room temperature, each dialysate was transferred to a 50 m l centrifuge tube and centrifuged (RC-2B Automatic Superspeed Refrigerated Centrifuge, Ivan Sorvall Inc., Norwalk, CT) at 8000 rpm for 15 min. 49 Each dialysate was analyzed i n t r i p l i c a t e and a mean value was calculated. For each dialysate, 3 ml aliquots of clear supernatant were transferred to each of 3 test tubes. Then 1.5 ml of chromagen solution were added. Each tube was mixed using a vortex mixer (Thermolyne Corp., Dubuque, I A ) and allowed to react 10 min prior to reading the absorbance of the solution at 533 nm, A Cary 210 spectophotometer (Varian Associates Inc., Palo Alto, CA) was used to determine absorbance and d.d. water was used to zero the instrument. For each run, three reagent blanks and two iron standards (0.5 ppm and 1.0 ppm iron in O.LN HCL) were treated and analyzed by the same procedures as the dialysates. A sample blank was prepared for each dialysate in order to account for any absorbance due to the color of the dialysate i t s e l f . For each dialysate, one 3 ml aliquot of clear supernatant was transferred to a test tube and 1.5 ml of d.d. water were added. After mixing, the absorbance of the sample blanks was read at 533 nm and the value subtracted from the absorbance for the bathophenanthroline reactive iron determined for each dialysate. A standard curve was prepared using iron standards made up in 0.1N HCL (see Appendix A). The equation of the line was determined by linear regression on a Monroe calculator (Litton Business Systems Inc.). The concentration of iron in each dialysate was determined using the standard curve and was calculated as ug Fe/ml dialysate. 50 4 . 2 . 8 Determination of nonheme iron As mentioned, the total iron content and nonheme iron content of a muffin product are essentially identical (Miller, 1982) . Duplicate samples of baked and unbaked muffin from each muffin batch were analyzed for total iron by atomic absorption spectroscopy. Frozen muffin samples were thawed at room temperature. Aliquots of thawed muffin were wet ashed by boiling with a mixture of hydrochloric and n i t r i c acids according to the method of Maurer (1977) ' The digested samples were cooled, made up to volume with d.d. water and f i l t e r e d (Whatman 5 4 l ) . The c l a r i f i e d solution was analyzed for iron by atomic absorption spectroscopy. Reagent blanks were carried through the entire ashing procedure. The concentration of iron in muffins was calculated on a dry weight basis. The total iron content of bran and unenriched flour was determined in t r i p l i c a t e by the same method. 4 . 2 . 9 Calculation of percent dialyzable iron The results were calculated as percent dialyzable iron (%Dl) according to the following equations % DI • A x B x 100 (1) CxD where; A B ug Fe/ml dialysate B = ml dialysate C • ug nonheme Fe/g muffin (dry weight basis) D = g muffin in 20 g aliquot pepsin digest (dry weight basis) 51 5- RESEARCH DESIGN AND ANALYSIS OF DATA Two preliminary studies were carried out to determine i f the results obtained by Miller et a l . (1981) using their in vitro method could be reproduced. Two test meals were formulated and prepared to duplicate those used by Miller et a l . (1981). The experimental results obtained were similar to those presented by Miller et a l . (1981) and indicated that the methodology could be used to estimate relative iron availability. The results of the preliminary studies are presented in Appendix B. Seven experiments were conducted to estimate the relative a v a i l a b i l i t y of the intrinsic iron of wheat bran when incorporated into a muffin product. Baked and unbaked muffins from each batch were blended to a slurry separately with either water, orange juice or aqueous solutions of ascorbic acid, c i t r i c acid or ascorbic and c i t r i c acids in amounts assumed to be provided by 250ml of orange juice. A summary of the conditions varied in each experiment is presented in Table LX. The s t a t i s t i c a l analysis was performed using an Amdahl 470^8 computer and a program adapted by Greig and Osterlin (1978). The data were analyzed by two factor analysis of variance and mean values of % dialyzable iron were compared by Duncan's multiple range test (Zar, 197^0. 52 Table IX. A summary of experimental conditions Experiment Bran Particle Size Slurry Composition 1 #10 130 g muffin 150 ml d.d. water 2 # 4 0 130 g muffin 150 ml d.d. water 3 #10 130 g muffin 150 ml d.d. water 130 mg ascorbic acid 4 #10 130 g muffin 250 ml orange juice 5 # 40 130 g muffin 250 ml orange juice 6 #10 130 g muffin 150 ml d.d. water 1.5 g c i t r i c acid 7 #10 130 g muffin 150 ml d.d. water 130 mg ascorbic acid 1.5 g c i t r i c acid 53 IV. RESULTS The data obtained from each of the experiments are summarized in Table X. Results of the analysis of variance and Duncan's multiple range test for the effects of acid treatment, baking and bran particle size on the a v a i l a b i l i t y of iron from wheat bran are presented in Tables XI, XII, and XIII respectively. The effect of bran particle size was not determined directly but was based on observations of significant differences in % DI between acid treatments which varied only in bran particle size. Superscripts are used to designate where or i f significant differences exist among mean values. Details of the analysis of variance and Duncan's multiple range test are discussed in Appendix G. Other experimental results which are not directly related to the discussion of iron a v a i l a b i l i t y from wheat bran are shown in Appendix D. Table X. Experimental data for the determination of % dialyzable iron (% Dl). Experiment3, Baking Muffin ug Fe % DI (Acid Treatment) Effect Batch g muffin (Duplicate) (dry weight) 1 Baked 1 67.36 t 0.21 0.007 + 0.008 2 69.67 - I.65 0.005 + 0.005 Unbaked 1 70.13 - 2.86 0.005 0.005 2 68.24 i 1.95 0.002 + 0.004 2 Baked 1 63.45 - 1.79 0.003 0.005 2 63.24 i 3-99 0.002 0.004 Unbaked 1 67.25 ± 2.02 0.002 0.004 2 65.78 ± 2.65 0.007 + 0.005 3 Baked 1 64.65 - 0.35 0.078 •+ 0.084 . 2 65.56 * 2.83 0.042 + 0.043 Unbaked 1 68.72 ± 0.74 0.017 + 0.027 2 68.87 i 2.55 0.067 0.053 4 Baked 1 66.53 - 0.73 4.213 + 0.217 2 64.23 ± 0.80 4.420 0.189 Unbaked 1 66.00 i 2.22 4.993 + 0.144 2 67.82 - 0.06 4.74S + 0.331 5 Baked 1 67.50 - 0.04 3.920 0.135 2 65.41 t 0.12 4.197 0.245 Unbaked 1 67.36 - O.09 4.445 t 0.172 2 64.88 - 1.71 4.927 - 0.285 Table X. (continued) 55 Baked 1 68.13 + 2.68 1.792 0.099 2 65.82 2.27 1.872 + 0.117 Unbaked 1 69.47 0.61 2.367 + 0.097 2 66.69 0.15 2.200 0.174 Baked 1 63.44 0.91 2.518 0.086 2 65.27 + 1.4? 2.255 ± O.O85 Unbaked 1 67.35 t 0.85 2.602 ± 0.117 2 66.22 + 0.71 2.672 O.067 ^ e f e r to Table LX for complete details of experimental conditions ^Values are expressed as means standard deviation, n=2. c + Values are expressed as means - standard deviation. The mean values were calculated from 6 values for % DI, 3 from each of 2 digests. Refer to Figure 2 for c l a r i f i c a t i o n of the experimental plan. 56 Table XI. The effect of acid treatment on % DI from wheat bran. Acid Treatment (Experiment #l) "Acid" Tested % DI b l 1 water 0.005 -0.006 h 2 water 0.003 -0.005 h 3 acorbic acid 0.051 t 0.058 h 4 orange juice 4.594 -0.374 P 5 orange juice 4.372 i 0.429 P 6 c i t r i c acid 2.057 -0.267 0 7 c i t r i c & ascorbic acids 2.512 i 0.182 k aRefer to Table IX for complete details of experimental conditions. ^Values are expressed as means t standard deviation, (n - 24) Q Means not sharing the same superscript are significantly different (p<0.05) Table XII. The effect of baking on % DI from wheat bran. Baking % DI3-' rb Baked muffins 1.809 + 1.759 h Unbaked muffins 2.075 2.009 ^  aValues are expressed as means - standard deviation, (n - 84) bMeans not sharing the same superscript are significantly different (p<0.05) 57 Table XIII. The effect of bran particle size on % DI from wheat bran. Experiment # Bran Particle Size % Dialyzable Iron ' 1 #10 0.005 - 0.006 h 2 #4o 0.003 - 0.005 h LL #10 4.594 t 0.374 P 5 #40 4.372 £ 0.429 P ^Values are expressed as means - standard deviation, (n - 24) ^Means not sharing the same superscript are significantly different (p < 0.05). 58 IV. DISCUSSION 1. INTERPRETATION OF RESULTS It was the purpose of this study to examine i f acid treatment, baking and/or bran particle size had any effect on the availability of iron from wheat bran. As in any s c i e n t i f i c analysis, the results of this study were influenced by experimental error. However, the s t a t i s t i c a l model (see Appendix C) was designed to take the va r i a b i l i t y due to experimental error into consideration in the determination of any significant treatment effects. As a result, the treatment effects themselves were found to significantly influence the ava i l a b i l i t y of iron from wheat bran in spite of the fact that significant variation could be attributed to experimental error in the determination of iron availability. It should be noted that the determination of s t a t i s t i c a l l y significant treatment effects does not necessarily indicate that they are of any practical significance. However, the results found in this study did seem reasonable and attempts have been made to discuss the experimental findings based on information provided in the literature. 2. AVAILABILITY OF IRON FROM WHEAT BRAN Results of Experiments 1 and 2, as shown in Table XI i l l u s t r a t e that essentially none of the iron from wheat bran was available for absorption when bran muffins were combined with water. The form of iron present in wheat bran is not known. Work carried out by Camire and Clydesdale (1982, 1981) suggested that the endogenous iron of bran exists in a water insoluble form and may be complexed with protein. Morris and 59 E l l i s (1982) recently determined that over 6O/0 of the endogenous iron in wheat bran is present as monoferric phytate (MFP) which is probably bound to protein and other cellular components. Morris and E l l i s (1982) investigated the av a i l a b i l i t y of iron fed as MFP using the hemoglobin repletion test in rats. It was found that the iron from purified MFP prepared from extracts of wheat bran and 2 synthetic preparations was highly available. Simpson et a l . (1981) measured the absorption of iron fed as free MFP using the extrinsic tag method in humans. The MFP was added to meals of both high and low av a i l a b i l i t y . It was determined that iron fed as MFP was no less well absorbed by human subjects than the dietary nonheme iron in the meals. However, in contrast to the reported av a i l a b i l i t y of iron fed as free MFP, i t is generally recognized that the a v a i l a b i l i t y of endogenous iron from wheat bran is poor (Turnlund, 1982; Erdman, 1981). Numerous studies have shown that when wheat bran was added to a meal the absorption of dietary iron was decreased (Morris and E l l i s , 1982; Reinhold, 1982; Reinhold et a l . 1981; Simpson et a l . , 198l). Although the nature of the inhibition i s unknown i t has been attributed to the formation of complexes of iron with such compounds as phytate and dietary fiber (Simpson et a l . , 1981). It seems reasonable to assume that the same binding mechanisms may be responsible for the complexation and poor a v a i l a b i l i t y of the endogenous iron of wheat bran. Therefore, the findings of some recent studies of the interaction of iron with wheat bran w i l l be discussed. Thompson and Weber (1979) investigated the binding of endogenous copper, zinc and iron i n 6 fiber sources, including wheat bran, using an in vitro method. A dual incubation procedure at pH O.65 and then at 60 pH 6.8 was used to simulate the changes in pH that occur during the in vivo digestion process. It was found that most minerals were released into solution after incubation at pH O.65 but were bound after incubation at pH 6.8. These results suggest that while the endogenous minerals of bran may not be bound at the very acidic pH of the stomach they may be rebound as the pH is raised and thereby become unavailable for absorption in the duodenum. Simpson et a l . (1981) investigated the effect of whole wheat bran and i t s components on the absorption of nonheme dietary iron using a double isotope technique in humans. In their f i r s t experiment, test meals consisted of 2 muffins and a milkshake. The muffins were either plain or contained 6 g of wheat bran. It was found that bran muffins reduced absorption of dietary nonheme iron by 7^% compared to absorption from a meal containing plain muffins. In their next experiment, Simpson et a l . (1981) examined the effect of bran phytate on iron absorption. They found that dephytinized bran produced the same degree of inhibition of dietary nonheme iron absorption as did whole bran and they suggested that the inhibitory effect of bran should not be attributed to i t s phytate content. As reported by Maga (1982), the binding of iron by phytate has been extensively studied but many of the data are contradictory. Some investigators have shown phytate to be inhibitory to iron absorption whereas others have shown i t to have no effect. In a third experiment, Simpson et a l . (1981) attempted to identify the fraction of wheat bran that inhibits iron absorption. Dephytinized bran was separated into a soluble, phosphate-rich fraction and an insoluble, high-fiber fraction. The total phosphorus content was 13 mg in insoluble 61 bran and 166 mg in soluble bran; neutral detergent fiber was 4.6 g in insoluble bran and was undetectable in the soluble fraction. It was found that inhibition was greater with soluble than with insoluble bran, suggesting that phosphate was more inhibitory than fiber. However, these workers concluded that they could not clearly identify the bran fraction responsible for the inhibition of iron absorption. In another study, Reinhold et a l . (1981) investigated the binding of iron by wheat bran and the dietary fiber of wheat bran using an in vitro method. Neutral detergent fiber (NDF), which contains cellulose, l i g n i n , hemicellulose and pectin; and acid detergent fiber (ADP), which is comprised of cellulose and lig n i n , were prepared from AACC wheat bran for use in the investigation. Native wheat bran, dephytinized wheat bran and cellulose were also examined in the study. The quantity of iron bound by wheat NDF was about O.38 ug of iron per g of NDF. Binding of iron was minimal, although appreciable, below pH 4.0, but increased rapidly above pH 5-0 to a maximum near pH 7.0. Reinhold et a l . (I98I) attempted to explain the binding of iron by wheat NDF. They observed that the uptake of iron with increasing pH produced a sinusoidal curve with i t s midpoint at about pH 5«8 which suggested the presence of a proton-dissociating group. It had been reported that t i t r a t i o n of wheat bran yielded a curve with an inflection near pH 6.0 and that iron binding by wheat gluten increased between pH 4.0 and 7'0' It was postulated that fiber-bound protein could be similarly active. However, removal of nearly a l l of the fiber-bound protein by treatment of NDF with peptidases failed to alter the pH dependence of iron binding. It was then suggested that the binding of iron by NDF could be due to the unsubstituted uronic acid groups of hemicellulose. 62 Acid detergent fiber of wheat was found to bind approximately 0.25 ug of iron per mg of ADF. The binding of iron by ADF was attributed to i t s cellulose component. Native bran behaved li k e NDF with respect to iron uptake and response to pH but had a lower a f f i n i t y for iron. Finally, dephytinized wheat bran was found to bind iron as effectively as native bran. It was concluded by Reinhold et a l . (1981) that the dietary fiber of wheat bran appears to bind dietary nonheme iron. It seems possible that dietary fiber may also have the a b i l i t y to bind the endogenous iron of bran. Camire and Clydesdale (1981) investigated the metal binding capacity of wheat bran and fractionated wheat bran from which water soluble polysaccharides, protein and starch had been removed. The binding of added ferrous sulfate was determined using an in vitro method. It was found that the fractionation treatment resulted in significantly more iron bound in comparison to the unfractionated wheat bran. It was suggested that this could be due to the fact that the lignin, cellulose and hemicellulose concentrations were increased as a result of the fractionation treatment. Results from the same study had indicated that lignin had a high metal binding capacity. As mentioned, the iron from wheat bran was found to be unavailable for absorption under the conditions of in vitro method used in the present study. It is postulated that at least some of the endogenous iron of wheat bran may exist in a form such as monoferric phytate which is complexed with protein components of the bran. During peptic digestion at acidic pH the iron may be released from bran in a soluble form. However, at neutral and alkaline pH the iron may be rebound to various components of the dietary fiber of wheat bran and made unavailable for 63 absorption. The iron i n wheat bran may also e x i s t i n a form which i s not so l u b i l i z e d during the digestion procedure and i s therefore not poten t i a l l y available f o r absorption. I t i s also possible that other ingredients of the muffin product such as egg or milk may bind any iron released from wheat bran. However, the effect of dietary components on the absorption of dietary nonheme iron i s a topic i n i t s e l f . Since a l l of the experiments u t i l i z e d the same muffin 'formulation, i t has been assumed that any possible effects of muffin ingredients were constant and therefore not of any significance i n the present study of r e l a t i v e iron a v a i l a b i l i t y . Clydesdale (1982) recently reviewed several proposed mechanisms for the binding of cations by the dietary f i b e r component, pectin, and suggested that the proposed mechanisms could have significance i n iron a v a i l a b i l i t y . A simple model postulated that a c i d i c polysaccharides such as pectin could be converted into anion exchangers by complexing with t r i v a l e n t cations, which have the a b i l i t y to bind a variety of anions. Another model, described i n d e t a i l by Clydesdale (1982), suggested that two carboxylic acid groups, which belong to neighboring galacturonic acid residues i n the l i n e a r chain of pectin, could participate i n binding • with one t r i v a l e n t cation. A t h i r d carboxylic acid group could be furnished by a neighboring l i n e a r chain and resul t i n complete neutralization of the t r i v a l e n t cation; thereby i n h i b i t i n g i t s cationic nature. I t was suggested that insoluble pectin fi b e r s which have been neutralized with a t r i v a l e n t cation would have a greater a f f i n i t y f o r various anions than a f i b e r which has been neutralized by a divalent cation, due to a greater density of positive charge. 64 Clydesdale (1982) suggested that such reactions could partially explain the greater ava i l a b i l i t y of ferrous iron compared to ferric iron. It was postulated that perhaps there is a tendency for trivalent ferric iron to form cationic complexes while divalent ferrous iron may tend to form neutral complexes. As a result, the reactivity of the ferric complex would increase, thereby increasing i t s potential for further binding and decrease i t s potential availability. Conversely, the ferrous complex would be soluble and more chemically inert so that i t would be more available for absorption. 65 3. EFFECT OF ORANGE JUICE ON IRON AVAILABILITY The effects of various acid treatments on the relative a v a i l a b i l i t y of iron from wheat bran, as estimated by % dialyzable iron (% Dl), are compared in Table XI. Among the acid treatments examined in the present study, orange juice showed the most significant (p< 0.05) enhancement of iron a v a i l a b i l i t y ; resulting in % DI values of approximately 4.59% and 4.37% from "meals" containing muffins made from # 10 and # 40 bran respectively. In order to evaluate the effect of orange juice on the a v a i l a b i l i t y of iron from wheat bran; ascorbic acid, c i t r i c acid and a combination of ascorbic and c i t r i c acids, in amounts assumed to be present in 250 ml of orange juice, were added to muffin "meals". The substitution of orange juice with a solution of ascorbic and c i t r i c acids resulted in a % DI value of 2.51%; while c i t r i c acid alone and ascorbic acid alone resulted in % DI values of about 2.06% and 0.051% respectively. The combined effect of c i t r i c and ascorbic acids was significantly (p< 0 . 0 5 ) less than the effect of orange juice, but was significantly (p <0.05) greater than the effect of c i t r i c acid alone. However, the a b i l i t y of c i t r i c acid to enhance iron a v a i l a b i l i t y was significantly (p< 0 . 0 5 ) greater than that of ascorbic acid alone. It is known that the a v a i l a b i l i t y of iron is related to the chemical state of iron. In a recent study, Kojima et a l . (1981) examined the effect of ascorbic acid, c i t r i c acid and orange juice on the solubilization of iron from cooked pinto beans using an in vitro method. Bean suspensions c were incubated with the test solutions at 37 C. After centrifugation of the mixtures, the supernatants were analyzed for iron. Although the study does not examine the a v a i l a b i l i t y of iron from wheat bran, i t does 66 provide insight into the effect of orange juice and i t s constituent organic acids on the solubilization of iron from a vegetable food. In a single incubation procedure a 10 mM solution of ascorbic acid was incubated with a cooked pinto bean suspension for 30 min at pH 2. Approximately 50% of the iron from the bean suspension was solubilized and virtu a l l y a l l of the soluble iron was in the ferrous state. In another procedure, 10 mM solutions of ascorbic acid were incubated with the bean suspension at variable pH. It was found that ascorbic acid was maximally effective in solubilizing iron in the pH range from 1.5 to 5' Throughout the pH range almost a l l of the supernatant iron was found in the ferrous form. Kojima et a l . (1981) postulated that the a b i l i t y of ascorbic acid to enhance the solubilization of iron from a cooked pinto bean suspension may occur via a reductive iron release mechanism. The ascorbic acid may be able to reduce tightly bound f e r r i c iron to the ferrous form with subsequent release of the iron from the ligands associated with the insoluble phase. The effect of c i t r i c acid on iron solubilization was also investigated. It was found that the a b i l i t y of a 10 mM solution of c i t r i c acid to solubize iron from a cooked pinto bean suspension was maximal near pH 6 and had l i t t l e i f any effect in the acidic range. A 10 mM solution of c i t r i c acid solubilized approximately k6% of the iron from a cooked pinto bean suspension and 25% of the soluble iron was in the ferrous state. A dual incubation procedure was used to examine the effect of combining c i t r i c acid and ascorbic acid. The bean suspension was f i r s t incubated at pH 2 for 30 min in the presence of a 10 mM solution of ascorbic acid. After the pH was raised to 6.0, a 10 mM solution of c i t r i c acid was 67 added and the mixture reincubated for an additional 30 min. It was found that when ascorbic and c i t r i c acids were combined their effects were additive and virt u a l l y a l l of the complexed iron was in the reduced form. Kojima et a l (1981) suggested that the solubilization of iron from a cooked pinto bean suspension by c i t r i c acid may be by chelation of iron in the f e r r i c form. The effect of various beverages on iron solubilization was examined using a dual incubation procedure. It was found that three orange juice preparations; Tang, frozen reconstituted, and fresh orange juice, were active in solubilizing iron from a cooked pinto bean suspension. Fresh orange juice solubilized approximately the same amount of iron as did a combination of 10 mM c i t r i c acid and 10 mM ascorbic acid in a dual incubation experiment. It was proposed that the a b i l i t y of orange juice to solubilize iron was most l i k e l y due to i t s content of c i t r i c and ascorbic acids. In another recent study, Reinhold et a l . (1981) investigated the binding of iron by the fiber of wheat bran using an in vitro method at pH 6.45. It was found that ascorbic acid and c i t r i c acid, among other compounds such as cysteine, EDTA and phytate, strongly decreased the binding of iron by wheat bran and by the neutral detergent fiber of wheat bran. Generally, inhibition of iron binding occurred at less than 1 mM/L and was proportional to the log of the concentration of the inhibitor. However, the behaviour of ascorbic acid was unique in that inhibition was small and constant at concentrations below 0.3 mM/L. However, when this concentration was exceeded, binding became proportional to the log of the concentration of ascorbic acid. It was concluded that among the inhibitors of iron binding examined, ascorbic acid was the least active 68 at low concentrations because of the plateau of inhibiting activity-shown by i t . Reinhold et a l . (1981) suggested that the a b i l i t y of ascorbic and c i t r i c acids to inhibit the binding of added iron by wheat bran indicates that these compounds may promote iron absorption by their a b i l i t y to release iron from i t s combination with dietary fiber. The exact mechanism by which ascorbic acid enhances the absorption of nonheme iron is not completely understood. Brise and Hallberg (1962) and Monsen and Page (1978) suggested that the effect of ascorbic acid was mainly due to i t s reducing action within the intestinal lumen. Ascorbic acid may increase the absorption of nonheme iron from foods by converting f e r r i c iron to the more available ferrous form. Other workers suggested that ascorbic acid forms a soluble chelate with f e r r i c iron and thereby prevents the formation of unabsorbable polymers and precipitates at neutral and alkaline pH (Conrad and Schade, I968; Crosby, 1968). In a recent review, Clydesdale (1982) proposed a mechanism for the effect of ascorbic acid on increasing the a v a i l a b i l i t y of nonheme iron. It was explained that at acidic pH, ferrous and f e r r i c ions are soluble and exist as hydrates, with the standard reduction potential of Fe J (aqj + e — ^ Fe (aq.) being +770 mv. In the presence of ascorbic acid, which has a standard reduction potential of +440 mv, the formation of ferrous iron w i l l take place spontaneously. At acidic pH, both ferrous-ascorbate and ferric-ascorbate complexes may also form. However, the overall effect of ascorbic acid at acidic pH i s to act as a reducing agent. This implies that in an acidic solution containing f e r r i c iron and ascorbic acid that the ferrous form w i l l predominate and f e r r i c -69 ascorbate complexes w i l l tend to destabilize. It was further explained that as the pH is raised, the hydrates of iron begin to lose protons and form their respective hydroxides which are increasingly insoluble with increasing pH. In basic solutions the standard reduction potential of Fe(OH)3(s)+e" > Fe(0H) 2(s) + (OH)" is -560 mv. Since most foods have a standard reduction potential of 4400 mv, the formation of f e r r i c hydroxide i s favored with and without ascorbic acid. Therefore, the effect of ascorbic acid as the pH is raised appears to be to maintain the fe r r i c form of iron i n solution by forming a ferric-ascorbate complex. A review of the literature reveals that there has been l i t t l e reference to the a b i l i t y of c i t r i c acid to enhance iron a v a i l a b i l i t y . However, i t i s recognized that c i t r i c acid i s an effective chelating agent (Lindsay, 1976). Any molecule or ion with an unshared electron pair can coordinate or complex with metal ions. Foods contain a variety of compounds (called chelating agents or ligands) which have the a b i l i t y to chelate metal ions (Furia, 1964). The steric location and number of groups on a ligand for combination with metal ions determines the characteristics of the metal complexes. For example, EDTA has the a b i l i t y to form 6 coordinating bonds with iron resulting in a stable chelate which may not liberate iron for absorption while, c i t r i c acid, which has 3 carboxylate groups to bind metal, forms an unstable complex that holds iron in solution at neutral and weakly alkaline pH (Benjamin, 1967). The formation of a chelate involves an equilibrium reaction, M + Y i|.-l'' > MY, where M is the metal ion, Y i s the coordinating agent and MY the metal complex. The rate at which a metal complex is formed is determined by the s t a b i l i t y or equilibrium constant of the metal complex which is given by the following equationt 70 (2-) As the s t a b i l i t y constant increases, more of the metal is complexed and the metal in the complex i s more tightly bound (Lindsay, 1976; Furia, 1964). The formation of a metal chelate i s also influenced by pH. Kojima et a l . (1981) found that c i t r i c acid was most effective in solubilizing iron from a cooked pinto bean suspension at pH 6.0 and had l i t t l e effect at pH 2. They suggested that the carboxylic acid groups of c i t r i c acid are protonated at acidic pH which interferes with their a b i l i t y to bind iron. As reported by Lindsay (1976), a gradual increase in pH allows dissociation of the carboxyl groups and enhances chelating efficiency. In foods, s t a b i l i t y constants alone are of limited value in predicting the effectiveness of a chelate because of interference by a variety of competing substances such as other metal ions, precipitants and other chelating agents (Furia, 1964). Hydroxyl ions w i l l compete with the metal ion for the chelating agent resulting in the formation of an insoluble hydroxide. As described by Furia (1964), this involves the competition of a complexing agent (Y) and a precipitating agent (X) in the reaction, MX f Y ••••••• MY + X. If the reaction is forced to the right and i f the metal complex (MY) i s soluble, then the precipitate (MX) w i l l dissolve. This w i l l occur when the s t a b i l i t y constant of the metal complex (MY) is greater than the reciprocal of the solu b i l i t y product of the precipitate (MX), when the components M,X and Y are present in equimolar concentrations. There is very l i t t l e information on the s t a b i l i t y constants and solubility products of various iron complexes as they exist in foods. 71 Knowledge of these values would aid in the understanding of the binding mechanisms and ava i l a b i l i t y of food iron. The pH dependence for the binding of iron by dietary fiber has been described by Thompson and Weber (1979). As well, the effect of pH on the action of ascorbic and c i t r i c acids has been discussed (Kojima et a l . , 1981; Lindsay, 1976). It should be noted that in the present study, each of the muffin slurries was adjusted to pH 2.0 prior to the pepsin digestion and each slurry was subsequently adjusted to approximately pH 7«0 during the dialysis of iron. Therefore, differences in pH among the acid treatments was not a factor in the determination of relative iron av a i l a b i l i t y . The experimental results obtained in the present study were not a l l as expected. The most surprising observation was the lack of any significant (p<0.05) effect of ascorbic acid on the relative a v a i l a b i l i t y of iron from wheat bran as determined by the in vitro method. When bran muffins were combined with an amount of ascorbic acid assumed to be present in orange juice (Experiment 3» Table XI), the resulting % DI value was not significantly (p<0.05) different than from muffins combined with only water (Experiments 1 and 2, Table XI). This appears to indicate that the observed significant (p <0.05) absorption enhancing effect of orange juice (Experiments 4 and 5» Table XI) was probably not due, in any great extent, to i t s ascorbic acid content. It is possible that the concentration of ascorbic acid was insufficient to enhance the ava i l a b i l i t y of iron from wheat bran. The apparent dependency of the efficacy of ascorbic acid on concentration has been observed (Clydesdale, 1982; Reinhold et a l . , 1981). 72 The a b i l i t y of ascorbic acid to enhance the avai l a b i l i t y of nonheme iron i s well established (Monsen, 1982). It is hypothesized by this author that ascorbic acid, present in adequate concentration, may enhance the av a i l a b i l i t y of iron from bran by i t s a b i l i t y to act as a reducing agent at acidic pH. During peptic digestion, protein-bound iron may be released from bran in the f e r r i c form. Then ascorbic acid present in the digest may reduce the iron to the more available ferrous form. It is also possible that iron may remain tightly bound as f e r r i c iron to components of wheat bran. Ascorbic acid may reduce the ferric form to the ferrous form and thereby release iron from i t s association with bran. The results from Experiment 6 (Table Xi) indicate that much of the observed effect of orange juice may be attributed to the action of i t s c i t r i c acid content. Orange juice contains approximately 10 times more c i t r i c acid than ascorbic acid (Nagy, 1978; Clements, 1964). Based on information presented in the literature, i t is suggested by this author that c i t r i c acid may enhance the a v a i l a b i l i t y of iron from wheat bran, as measured by the in vitro method, by i t s a b i l i t y to chelate f e r r i c iron at alkaline pH. For example, the s t a b i l i t y constant of the ferric-citrate complex may be greater than the reciprocal of the s o l u b i l i t y products of various iron-fiber complexes. As a result, chelation by c i t r i c acid could liberate f e r r i c iron from bran in a soluble, available form. Results from Experiment 7 (Table XI) show that ascorbic and c i t r i c acids combined appeared to have an effect greater than the sum of the individual acids. Kojima et a l . (1981) demonstrated a similar effect on iron solubilization but they combined equal concentrations of ascorbic and c i t r i c acids, which is not representative of the actual acid 73 content of orange juice. In the present study, the addition of ascorbic and c i t r i c acids to muffin "meals" in amounts assumed to be present in orange juice, enhanced iron a v a i l a b i l i t y from bran, as measured by the in vitro method, to only half the extent of combining muffins with orange juice. It is possible that either the other organic acids present in orange juice, such as oxalic and malic acids, and/or some other component of the juice may also be responsible for enhancing the availability of iron from wheat bran. 4. EFFECT OF BAKING ON IRON AVAILABILITY The heat treatment received during baking was found to significantly (p<0.05) decrease the relative a v a i l a b i l i t y of iron from wheat bran, as determined by the in vitro method, when compared to the a v a i l a b i l i t y of endogenous iron from unbaked muffin products. The effect of baking on % dialyzable iron (% D l ) is shown in Taile XII. As outlined in Appendix C, the baking effect was determined to be essentially the same for each acid treatment. As has been discussed, previous studies have found baking to have l i t t l e effect on the a v a i l a b i l i t y of dietary nonheme iron. However, other recent studies have shown that processing changes the chemical form of iron present in some foods (Lee and Clydesdale, 1981; Lee and Clydesdale, 1980a; Lee and Clydesdale, 1980b; Lee and Clydesdale, 1979b). Changes in the chemistry of iron may be important since the chemical form of iron is known to influence iron a v a i l a b i l i t y . Lee and Clydesdale (1979b) developed a method for the simultaneous analysis of the chemical forms of iron, including elemental, soluble, complexed and ionic, added to or endogenous to foods. In one study, 74 Lee and Clydesdale (1980a) investigated the effects of baking on the chemical forms of iron in biscuits. Almost a l l of the endogenous iron of unenriched wheat flour was i n i t i a l l y found to be either in the insoluble or f e r r i c form. After the unenriched flour was baked in biscuits, the distribution of iron forms was measured. It was found that there was an increase in insoluble iron with a corresponding decrease in fe r r i c iron. The effect of baking on the chemical form of various iron enrichment sources was also investigated. Each of the iron sources was added to unenriched flour, baked as biscuits and the resulting "iron profile" measured. It was found that very large changes in the iron profile resulted from the baking process. The major effect of baking was the production of insoluble forms of iron; even when a highly soluble form of iron, such as ferrous sulfate, was used for enrichment. Camire and Clydesdale (1981) investigated the effect of wet and dry processing treatments on the a b i l i t y of AACC wheat bran and major fractions of dietary fiber to bind metals, including iron, in an in vitro system at pH 5.0 to 7«0. It was found that toasting wheat bran at 177°C for 1 h caused a highly significant increase in the amount of added ferrous iron bound by bran. In an earlier study, Anderson and Clydesdale (1980b) had examined the effect of processing on the dietary fiber content of AACC wheat bran. Their results showed that toasting wheat bran significantly increased the apparent amount of "lignin" in the bran. The content of "lignin" increased from 2.87% at time 0 to 4.28% after 30 min and 12.02% after 60 min of toasting at 177°C. An overall increase in dietary fiber found after toasting indicated to these workers that the increase in "lignin" content was due to fractions of the wheat bran other than the dietary 75 fiber constituents. It had. been previously reported that heating above 50°C was found to cause the formation of Maillard browning products which were insoluble in 72% sulphuric acid and therefore isolated along with the true l i g n i n fraction. Based on the work by Anderson and Clydesdale (1980), Camire and Clydesdale (I98I) concluded that the increased binding of some metals by wheat bran as a result of toasting could be due to the small increase in the amount of "lignin" formed during the heating process. Lignin and other components of dietary fiber have been shown to have high metal binding capacities (Clydesdale, 1982; Camire and Clydesdale, I98I). Based on the information reported by other workers, i t is postulated by this author that the baking of muffins may have resulted in a significant (p<0.05) decrease in the relative a v a i l a b i l i t y of iron from bran, as measured by % DI, due to the formation of substances during the baking process which were inhibitory to iron absorption. These substances may be similar to the dietary fiber component, lignin, and have the a b i l i t y to complex iron in an insoluble form. However, i t i s recognized that numerous factors interact to affect the ava i l a b i l i t y of dietary nonheme iron. Therefore, i t i s probable that the observed effect of baking on the relative a v a i l a b i l i t y of iron from bran is of l i t t l e practical importance compared to the influence of other dietary constituents, such as orange juice, consumed at the same time as the baked product. 76 5. EFFECT OF BRAN PARTICLE SIZE ON IRON AVAILABILITY As shown in Table XIII, a decrease in the particle size of bran had no significant (p<0.05) effect on the ava i l a b i l i t y of iron from wheat bran, as determined by the in vitro method. There was no significant (p<0,05) difference in dialyzable iron between pairs of experiments in which the only factor varied was bran particle size (Table LX). It was hypothesized by this author that a decrease in bran particle size would increase the exposure of wheat bran to the digestive process and increase the liberation of endogenous iron from bran. However, even i f more iron was liberated from finely ground bran (# 40) during digestion, the iron may not have been available for absorption. As has been discussed, chemical and dietary factors in the gut influence the ava i l a b i l i t y of dietary nonheme iron. It appears that under the conditions of this study, bran particle size is not a factor in the av a i l a b i l i t y of iron from wheat bran. 77 6. LIMITATIONS OF THE STUDY In the present study, the ava i l a b i l i t y of iron from wheat bran was estimated using an in vitro method. The method simulated gastro-intestinal digestion but did not account for such factors as active transport or brush border binding proteins which may play a role in the absorption of food iron in vivo. Only soluble, low molecular weight iron, capable of passing through a membrane by simple diffusion, was measured as an estimate of available iron. Therefore, because of the uncertainties involved with the determination of iron availability using an a r t i f i c i a l system, the results from the present study represent the relative a v a i l a b i l i t y rather than the absolute av a i l a b i l i t y of iron from wheat bran. No conclusions regarding the absorption of iron from bran in vivo can be made. The actual absorption and u t i l i z a t i o n of food iron can only be measured using a biological system. However, comparison of results u t i l i z i n g the in vitro method of Miller et a l . (1981) and human in vivo methods for the determination of iron a v a i l a b i l i t y has shown good agreement between the methods (Schricker et a l . , I 9 8 I ) . For this reason, the observed effects of such factors as orange juice and baking on the avail a b i l i t y of iron from bran in vitro probably approximates their effects in the human body and provide a basis for further investigation of iron a v a i l a b i l i t y from wheat bran. 78 VI. CONCLUSIONS There has been l i t t l e information published on the a v a i l a b i l i t y of the endogenous iron of wheat bran. In this study, the in vitro method developed by Miller et a l . (1981) was found to provide a relatively rapid and inexpensive estimation of the amount of iron available for absorption from wheat bran. The method was based on the assumption that iron i s absorbed as soluble, low molecular weight complexes capable of passing through a membrane by simple diffusion. The endogenous iron from wheat bran was found to be virtually unavailable for absorption when the muffins were combined with water. On the basis of results obtained, i t is postulated that iron released from wheat bran during peptic digestion at acidic pH i s rebound to components of dietary fiber at neutral and alkaline pH. Orange juice was found to significantly increase the ava i l a b i l i t y of iron from wheat bran. The c i t r i c acid present in orange juice was determined to be at least partly responsible for the a b i l i t y of orange juice to promote iron absorption. It i s postulated that c i t r i c acid enhances the a v a i l a b i l i t y of iron by chelating with f e r r i c iron at neutral and alkaline pH and releasing i t from wheat bran in a soluble form. Although ascorbic acid has been shown to increase the a v a i l a b i l i t y of nonheme iron, i t appears that the effect of ascorbic acid is dependent on concentration. The concentration of ascorbic acid used in this study appeared to be inadequate for the promotion of iron ava i l a b i l i t y from wheat bran. However,when ascorbic and c i t r i c acids were combined in amounts assumed to be present in orange juice they appeared to have an effect 79 greater than the sum of the individual acids. Ascorbic acid may increase the avai l a b i l i t y of iron from wheat bran in conjunction with c i t r i c acid by i t s a b i l i t y to reduce f e r r i c iron to the more soluble ferrous state and/or i t s a b i l i t y to chelate f e r r i c iron and form soluble complexes at neutral and alkaline pH. Baking was found to significantly decrease the ava i l a b i l i t y of endogenous iron from wheat bran. It i s postulated that the heat treatment provided during baking results in the formation of insoluble forms of iron which are unavailable for absorption. Finally, bran particle size was found not to be a factor in the avai l a b i l i t y of iron from wheat bran. In conclusion, although the results of this study have provided some insight into the a v a i l a b i l i t y of iron from wheat bran, further research i s necessary to determine the chemistry of iron and iron binding. A clearer understanding of the chemical nature of iron in whaet bran and tha influence of individual and interacting factors on the chemical behavior of iron i n the gut may eventually lead to improved availabilty of the essential element iron from wheat bran. 80 REFERENCES AACC. 1978. "Approved Methods of the American Association of Cereal Chemists," American Association of Cereal Chemists, St. Paul, MN. Anderson, N.E. and Clydesdale, F.M. 1980a. An analysis of the dietary fiber content of a standard wheat bran. J. Pood Sci. 45: 336. Anderson, N.E. and Clydesdale, F.M. 1980b. Effects of processing on the dietary fiber content of wheat bran, pureed green beans, and carrots. J. Food Sci. 45: 1533. AOAC. 1980. " O f f i c i a l Methods of Analysis," 13th ed. Association of O f f i c i a l Analytical Chemists, Washington, DC. Benjamin, B.I., Gortell, S. and Conrad, M.E. I967. Bicarbonate-induced iron complexes and iron absorption: one effect of pancreatic secretions. Gastroenterology. 53j 389. Beutler, E. 1980. Iron. In "Modern Nutrition in Health and Disease," 6th ed. R.S. Goodhart and H.E. Shils(eds.), p. 324. Lea and Febiger, Philadelphia, PA. Bibeau, T.C. and Clydesdale, F.M. 1976. Availability, use and interaction of iron in food. Food Prod. Dev. 8: 130. Bjorn-Rasmussen, E. 1974. Iron absorption from wheat bread. Influence of various amonts of bran. Nutr. Metabol. 16: 101. Brise, H. and Hallberg, L. 1962. Effect of ascorbic acid on iron absorption. Acta. Medica. Scand. 171(Supl. 376): 51« Burkitt, D. 1977• Food fiber. Benefits from a surgeon's perspective. Cereal Poods World. 22: 6. Camire, A.L. and Clydesdale, F.M. 1981. Effect of pH and heat treatment on the binding of calcium, magnesium, zinc, and iron to wheat bran and fractions of dietary fiber. J. Food Sci. 46: 548. Camire, A.L. and Clydesdale, F.M. 1982. Interactions of soluble iron with wheat bran. J. Food Sci. 47: 1296. Clark, L.J. 1962. Iron(ll) determination in the presence of i r o n ( l l l ) using 4,7-diphenyl-l,10-phenanthroline. Analytical Chem. 34: 348. Clemens, R.A. and Mercurio, K.C. I98I. Effects of processing on the bioavailability and chemistry of iron powders in a li q u i d milk-based product. J. Food Sci. 46: 30. Clements, R.L. 1964. Organic acids in citrus f r u i t s . 1. Varietal differences. J. Food Sci. 29: 276. 81 Clydesdale, F.M. 1982. The effects of physiochemical properties of food on the chemical satus of iron. In "Nutritional Bioavailability of Iron," C. Kies (ed.), p. S5- ACS Symposium Series 203, American Chemical Society, Washington, DC. Coccodrilli, Jr., G.D., Reussner, G.H. and Thiessen, Jr., R. 1976. Relative biological value of iron supplements in processed food products. J. Agric. Food Chem. 24s 351. Conrad, M.E. and Schade, S.G. I968. Ascorbic acid chelates in iron absorption: a role for hydrochloric acid and b i l e . Gastroenterology. 55: 35. Cook, J.D. 1978. Food iron a v a i l a b i l i t y . Food and Nutrition News, National Livestock and Meat Board. 49: 1. Crosby, W.H. 1968. Control of iron absorption by intestinal luminal factors. Am. J. Clin. Nutr. 21: II89. Disler, P.B., Lynch, S.R., Charlton, R.W., Bothwell, T.H., Walker, R.B. and Mayet, F. 1975' Studies on the f o r t i f i c a t i o n of cane sugar with iron and ascorbic acid. Br. J. Nutr. 34: 1^1• Elwood, P.C., Newton,D., Eakins, J.D. and Brown, D.A. I968. Absorption of iron from bread. Am. J. Clin. Nutr. 21: 1162. Erdman, Jr., J.W. 1981. Bioavailability of trace minerals from cereals and legumes. Cereal Chem. 58: 21. Fritz, J.C. and Fla, G.W. 1972. Application of the animal hemoglobin repletion test to measurement of iron a v a i l a b i l i t y in foods. J. AOAC. 55: 1128. F r i t z , J.C, Pla, G.W., Harrison, B.N. and Clark, G.A. 1974. Collaborative study of the rat hemoglobin repletion test for bioavailability of iron. J. AOAC. 57: 513. F r i t z , J.C, Pla, G.W., Harrison, B.N., Clark, G.A. and Smith, E.A. 1978. Measurement of the bioavailability of iron using the rat hemoglobin repletion test. J. AOAC. 6ls 709. Furia, T.E. 1964. EDTA in foods. Food Technol. 17: 50. Greig, M. and Osterlin, D. 1978. "UBC Anovar. Analysis of Variance and Covariance." Computing Center, The University of B r i t i s h Columbia, Vancouver, B.C. Hallberg, L- 1980. Food iron absorption. In "Iron," J.D. Cook(ed.), p. 116. Churchill Livingstone, New York, NY. 82 Hart, H.V. 1971. Comparison of the ava i l a b i l i t y of iron in white bread f o r t i f i e d with iron powder with that of iron naturally present in wholemeal bread. J. Sci. Food Agric. 2 2 : 354. Hazell, T., Ledward, D.A. and Neale, R.J. 1978. Iron ava i l a b i l i t y from meat. Br. J. Nutr. 3 9 : 631. Health and Welfare Canada. 1977• "Recommendations for Prevention Programs in Relation to Nutrition and Cardiovascular Disease," Health and Welfare Canada, Ottawa. Health and Welfare Canada. 1979. "Nutrient Value of Some Common Foods," Health and Welfare Canada, Ottawa. Health and Welfare Canada. 1981. "The Food and Drugs Act and Regulations," with amendments to March 19, I 9 8 I . Queen's Printer and Controller of Stationery, Supply and Services Canada, Ottawa. Hussain, R., Walker, R.B., Layrisse, M., Clark, P. and Finch, CA. 1965. Nutritive value of food iron. Am. J. Clin. Nutr. l 6 » 464. Innis, S. I98I. Lecture material, HUNU 517• School of Home Economics, University of B r i t i s h Columbia, Vancouver, B.C. Jacobs, A. and Greenman, D.A. 1969« Availability of food iron. Br. Med. J. 1 : 673. Kelsay, J.L. 1978. A review of research on effects of fiber intake on man. Am. J. Clin. Nutr. 3 1 : 142. Kent, N.L. 1975- "Technology of Cereals," 2nd ed. Pergamon Press Ltd., Oxford. Kies, C.(ed.) I 9 8 2 . "Nutritional Bioavailability of Iron," ACS Symposium Series 203, American Chemical Society, Washington,DC. Kojima, N., Wallace, D. and Bates, G.W. I 9 8 I . The effect of chemical agents, beverages, and spinach on the i n vitro solubilization of iron from cooked pinto beans. Am. J. Clin. Nutr. 34: 1392. Kuhn, I.N., Layrisse, M., Roche, M., Martinez,C. and Walker, R.B. 1968. Observations on the mechanism of iron absorption. Am. J. Clin. Nutr. 2 1 : 1184. Layrisse, M. and Martinez-Torres, C. 1971. Food iron absorption: iron supplementation of food. Prog. Hematol. 7: i37« Lee, K. 1982. Iron chemistry and bioavailability in food processing. In "Nutritional Bioavailability of Iron," C.Kies(ed.), p. 27. ACS Symposium Series 2 0 3 , American Chemical Society, Washington, DC. 83 Lee, K. and Clydesdale, F.M. 1979a. Iron sources used in food f o r t i f i c a t i o n and their changes due to food processing. CRC C r i t i c a l Reviews in Food Science and Nutrition, l i s 117* Lee, K. and Clydesdale, F.M. 1979b. Quantitative determination of the elemental, ferrous, f e r r i c , soluble, and complexed iron in foods. J Food Sci. 4 4 : 549. Lee, K. and Clydesdale, F.M. 1980a. Effect of baking on the forms of iron in iron-enriched flour. J. Food Sci. 4 5 : 1500. Lee, K. and Clydesdale, F.M. 1980b. Chemical changes of iron in food and drying processes. J. Food Sci . 4 5 : 711» Lee, K. and Clydesdale, F.M. I98I. Effect of thermal processing on endogenous and added iron in canned spinach. J. Food Sci. 46: 1064. Lee, G.F. and Stumm, W. i960. Determination of ferrous iron in the presence of f e r r i c iron with bathophenanthroline. J. Amer. Water Works Assoc. 5 2 : I567. Leichter, J. and Joslyn, M.A. 1967. The state of iron in flour, dough and bread. Cereal Chem. 44: 346. Lindsay, R.C. 1976. Other desirable constituents of foods. In "Food Chemistry," O.R. Fennema(ed.), p. 465. Marcel Dekker, Inc., New York, NY. Lock, S. and Bender, A.E. I98O. Measurement of chemically-available iron in foods by incubation with human gastric juice in vit r o . Br. J. Nutr. 4 3 : 413. Maga, J.A. I982. Phytate; Its chemistry, occurrence, food interactions, nutritional significance, and methods of analysis. J. Agr. Food Chem. 30*1 Mahoney, A.W. and Hendricks, D.G. 1982. Efficiency of hemoglobin regeneration as a method of assessing iron bioavailability in food products. In "Nutritional Bioavailability of Iron," C. Kies(ed.), p . l . ACS Symposium Series 203, American Chemical Society, Washington, DC. Maurer, J. 1977• Extraction method for the simultaneous determination of Na, K, Ca, Mg, Fe, Cu, Zn and Mn in organic material using AAS. Lebensmittel-Untersuchung und-Forschung. 1 6 5 : 1. Miller, D.D. I982. Personal communication. Miller, D.D. and" Schricker, B.R. I982. In vitro estimation of food iron bioavailability. In "Nutritional Bioavailability of Iron," C.Kies(ed.), p. 11. ACS Symposium Series 203, American Chemical Society, Washington,DC. 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. Monsen, E.R. 1982. Ascorbic acid: An enhancing factor in iron absorption. In "Nutritional Bioavailability of Iron," C. Kies(ed.), p. 85. ACS Symposium Series 203, American Chemical Society, Washington, DC. Monsen, E.R. and Page, J.F. 1978. Effects of EDTA and ascorbic acid on the absorption of iron from an isolated rat intestinal loop. J. Agric. Food Chem. 26: 223. Monsen, E.R., Hallbeorg,!., Layrisse, M., Hegsted, D.M., Cook, J.D., Mertz, W., and Finch, CA. 1978. Estimation of available dietary iron. Am. J. Clin. Nutr. 31: 134. Morck, T.A. and Cook, J.D. 1981. Factors affecting the bioavailability of dietary iron. Cereal Foods World. 26: 667. Morris, E.R. and E l l i s , R. 1982. Phytate, wheat bran, and bioavailability of dietary iron. In "Nutritional Bioavailability of Iron," C. Kies(ed.), p. 121. ACS Symposium Series 203, American Chemical Society, Washington, DC Nagy, S. I978. Fruit juices. In "Encyclopedia of Food Science," M.S. Peterson and A.H. Johnson(eds.), p.334. AVI Publishing Co., Westport, CT. Narasinga-Rao, B.S. and Prabhavathi, T. 1978. An in vitro method for predicting the bioavailability of iron from foods. Am. J. Clin. Nutr. 31: I69. Narins, D. 1980. Absorption of nonheme iron. In "Biochemistry of Nonheme Lron," A.. Bezkorovainy(ed.), p.47. Plenum Press, New York, NY. Nojeim, S.J. and Clydesdale, F.M. I98I. Dissolved oxygen concentration and iron valence in a model system. J. Food Protection. 44: 762. O'Dell, B.L., deBoland, A.R. and Koirtyohann, S.R. 1972. Distribution of phytate and nutritionally important elements among the morphological components of cereal grains. J. Agric. Food Chem. 20: 718. Peters, Jr., 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 dia l y s i s . Gastroenterology. 6l: 315» Pla, G.W., Harrison, B.N. and F r i t z , J.C. 1973' Comparison of chicks and rats as test animals for studying bioavailability of iron, with special reference to use of reduced iron in enriched bread. J. AOAC 56: 1369. 85 Polizzoto, L.M., Tinsley, A.M., Weber, G.W. and Berry, J.W. 1983. Dietary fiber in muffins. J. Food Sci . 48: 111. Ranhotra, G.S, Lee, C., and Gelroth, J.A. 1979* Bioavailability of iron in some commercial variety breads. Nutr. Rept. Intl. 19: 851. Reinhold, J.G. i 9 8 2 . Dietary fiber and the bioavailability of iron. In "Nutritional Bioavailability of Iron," G. Kies(ed.), p. 143. AGS Symposium Series 2 0 3 , American Chemical Society, Washington, DC. Reinhold, J.G., Garcia, J.S. and Garzon, P. 1981. Binding of iron by fiber of wheat and maize. Am. J. Clin, Nutr. 34: 1384. Rossander, L., Hallberg, L. and Bjorn-Rasmussen, E. 1979- Absorption of iron from breakfast meals. Am. J. Clin. Nutr. 32: 2484. Sayers, M.H., Lynch, S.R., Jacobs, P., Charlton, R.W., Bothwell, T.H., Walker, R.B. and Mayet, F. 1973* The effects of ascorbic acid supplementation on the absorption of iron in maize, wheat and soya. Br. J. Haematology. 24: 209. Schackleton,L. and McCance, R.A. 1936. The ionizable iron in foods. Biochem. J. 30:582. cited by Hart ( l 9 7 l ) . Schricker, B.R. and Miller, D.D. 1982. In vitro estimation of relative iron a v a i l a b i l i t y in breads and meals containing different forms of fo r t i f i c a t i o n iron. J. Food Sci. 47: 723. Schricker, B.R., Miller, D.D., Rasmussen, R.R. and Van Campen, D. I98I. A comparison of in vivo and in vitro methods for determining av a i l a b i l i t y of iron from meals. Am. J. Clin. Nutr. 34: 2257. Simpson, K.M., Morris, E.R. and Cook, J.D. I 9 8 I . The inhibitory effect of bran on iron absorption in man. Am. J. Clin. Nutr. 3 4 : 1469. Smith, G.F., McCurdy, Jr. W.H. and Diehl, H. 1952. The colorimetric determination of iron in raw and treated municipal water supplies by use of 4:7-diphenyl-l:10-phenanthroline. Analyst. 77 : 4 l 8 . S p i l l e r , G.A. and Amen, R.J. 1974. Role of dietary fiber in nutrition. Food Prod. Dev. 1 0 : 30 . Subcommittee on Iron. 1979» Iron metabolism in humans and other mammals. In "Iron" p. 7 9 . University Park Press, Baltimore, MD. Theuer, R.C, Kemmerer, K.S., Marin, W.H., Zoumas, B.L. and Sarett, H.P. I97I. Effect of processing on av a i l a b i l i t y of iron salts in liquid infant formula products. J. Agric. Food Chem. 1 9 : 555. Theuer, R.C., Martin, W.H., Wallander, J.F. and Sarett, H.P. 1973« Effect of processing on a v a i l a b i l i t y of iron salts in liquid infant formula products. J. Agr. Food Chem. 2 1 : 482. 86 Thompson, S.A. and Weber, G.W. 1979. Influence of pH on the binding of copper, zinc and iron in six fiber sources. J. Food Sci. 4 4 : 752. Turnlund, J.R. 1982. Bioavailability of selected minerals in cereal products. Cereal Foods World. 2 7 ( 4 ) : 152. USDA. 1975« "Composition of Foods, Agriculture Handbook 8. " US Government Printing Office, Washington, DC. Verma, R.S., Motzok, I., Chen, S-S., Rasper, J. and Ross, H.U. 1977. Effect of storage in flour and of particle size on the bioavailability of elemental iron powders for rats and humans. JAOAC. 6 0 : 759« Waddell, J. 1974. Bioavailability of iron sources. Food Prod. Dev. 10s 80. Wood, R.J., Stake, P.E., Eisemand, J.H., Shippee, R.L., Wolski, K.E. and Koehn, U. 1978. Effects of heat and pressure processing on the relative biological value of selected dietary supplemental inorganic iron salts as determined by chick hemoglobin repletion assay. J. Nutr. 108: 1477-Zar, J.H. 1974. "Biostatisical Analysis,"Prentice Hall Inc., Englewood C l i f f s , N.J. 87 APPENDIX A STANDARD CURVE FOR DETERMINATION OF DIALYZABLE IRON A standard curve was prepared using 0 .05. 0.10, 0.20, 0 .30, 0.40, 0 .50, 0 .60, 0 .70, 0.80, 0 .90, 1.0, 1 .6 , 2.0;and 3.0 ppm iron standards in 0.1N HCL. Determinations of absorbance were made in replicates of 6. The data for the standard curve were analyzed by linear regression and yielded the following information: X 0.810 Y 0.146 P 0.181 * 0.000 S y x 0.005 r O.999 Y = mean of Y values /3 = slope of the best f i t regression line • Y intercept S y x = standard error of estimate r * correlation coefficient Given the general equation for a straight line as: the equation of the line for the standard curve was calculated as: Y = 0.000 + 0.181 X where: X mean of X values 88 APPENDIX B PRELIMINARY STUDIES Table I. Composition of test meals Meal Component Weight (g) 1 Ground beef, cooked 35.0 Bread, white enriched 19-5 Green beans, frozen 23.25 Fluid whole milk 85.O Water 87.25 2 Ground beef, cooked 35-0 Bread, white enriched 19.5 Green beans, frozen 23.25 Orange juice, frozen, reconstituted 85.0 Water 87.25 Table II. Total iron and % dialyzable iron (%DI) determined in the test meals. Meal Nonheme mg/100 g Iron meal % D 1* b c Experimental Published Experimental Published 1 0.5 0.7 3.78 i 0.12 4.08 * 0.31 2 0.5 0.8 19.52 - O.69 24.96 t 0.83 Values are expressed as means £ standard deviation, n«6. Calculated from tabular values (Health and Welfare Canada, 1979). CAs reported by Miller et a l . (1981). 89 APPENDIX C STATISTICAL ANALYSIS The data were analyzed by two factor analysis of variance (ANOV) with subsampling. Such an analysis is also called a nested or hierarchial ANOV.- This design is used when each group of data consists of subgroups and results in a complex partitioning of va r i a b i l i t y . Percent dialyzable iron (?aDl) was measured in 7 different experiments. Within each experiment there were 2 muffin batches. Analysis of 2 muffin batches duplicated the experiment. Within each muffin batch, % DI was determined in both baked and unbaked muffins. In the determination of % DI, the digestion procedure was duplicated within each baking effect. From each digest, 3 dialysates were collected and analyzed for dialyzable iron. The following s t a t i s t i c a l model was developed: Fe = A + B + AB + D(A) + BD(A) + C(ABD) f E where: A = the effect of acid treatment. Acid treatment refers to either water, orange juice, ascorbic acid, c i t r i c acid or a combination of ascorbic and c i t r i c acids. There are 7 acid treatments as defined by the 7 experiments. B = the effect of baking. There are 2 baking treatments (baked and unbaked) within each muffin batch. AB = the interaction effect of acid treatment and baking. C = pepsin digest. There are 2 aliquots taken for pepsin digestion within each baking treatment. D = duplicate. Each of the 2 muffins batches within each experiment is referred to as a duplicate. D(A) = duplicate error term. This term considers the error due to v a r i a b i l i t y between duplicates within each acid treatment. 90 BD(A) = consistency error term. This term measures the var i a b i l i t y among duplicates within the baking treatments. C(ABD) • digest error term. This term considers ths error due to differences between digests. E = residual error. The expected mean squares (EMS) table and ANOV are shown in Tables I and II respectively., The purpose of the ANOV was to determine i f the acid treatment and/or baking treatment had a significant effect on the avai l a b i l i t y of iron from wheat bran. The ANOV also determined i f there was an interaction between the 2 treatments. Sources of variation other than the effects of acid treatment or baking could influence the experimental results. However, the s t a t i s t i c a l model was designed to take the error due to var i a b i l i t y between duplicate determinations and/or the v a r i a b i l i t y between digests into consideration in the determination of any significant treatment effects. The effect of acid treatment on % DI was tested against the duplicate error term. The ANOV (Table II) indicated that there was a very significant difference between duplicate determinations. However, this source of variation was taken into consideration and a significant difference in % DI due to acid treatment was s t i l l found. The effect of various acid treatments was analyzed by comparing mean % DI of baked and unbaked samples combined for each acid treatment by Duncan's multiple range test (Zar, 197^). The effect of baking on % DI was tested against the consistency error term. Although the ANOV (Table Ii) showed that significant error could be attributed to va r i a b i l i t y among duplicates within the baking treatments, the effect of baking was s t i l l found to be significant. 91 Table I. Expected mean squares table. Source # Source EMS Tested Against 1 A s 2 4 1,4,6 4 2 B s 2 f 2,5,6 5 3 AB s 2 + 3,5,6 5 4 D(A) s 2 + 4,6 6 5 BD(A) s 2 + 5,6 6 6 G ( A B D ) 2 , s + 6 7 2 7 E s Table II. Analysis of variance Source DF Mean Square F Value F Probability A 6 97-45 731-59 0.0000 B 1 2.98 32.67 0.0009 AB 6 0.47 5.17 0.0251 D 7 0.13 8.92 0.0000 BD 7 0.09 6.10 0.0002 C 28 0.01 0.78 O.7697 ERROR 112 0.02 TOTAL 167 92 The effect of the baking treatments was examined by comparing mean % DI of a l l baked samples in a l l acid treatments to mean % DI of a l l unbaked samples in a l l treatments by Duncan's multiple range test (Zar, 1974). The v a r i a b i l i t y due to differences among digests did not directly influence the determination of significant treatment effects but was taken into consideration indirectly. As shown by ANOV (Table II), the interaction effect of acid treatment and baking was found to be significant. As illustrated in Figure 1, the significant interaction occurs only at the extremely low levels of % DI. It appears that the baking effect i s constant for each acid at more measurable levels of dialyzable iron. 93 Figure 1. A plot of the interaction of acid treatment and baking effect. 94 APPENDIX D ANALYSIS OF SAMPLES Table I. Particle size distribution of two bran fractions showing percentage of sample retained. o • a Sieve number Sieve mesh % Retained # 10 Bran # 40 Bran 18 1.00 mm 71.63 40 420 u 28.37 -60 246 u - 5 8 . 8 8 80 177 u - 32.89 120 125 u - 6 . 7 0 20 74 u - 1.52 a T y l e r test sieves. Table II. Average moisture content of flour, bran and muffins. Sample Replicates % Moisture f l o u r 3 3 11.90 - 0.07 AACC bran 3 9.70 * 0 .06 # 10 bran 3 9.78 - 0.07 # 40 bran 3 9.51 - 0.14 Baked muffins 30 30.98 - 1.10 Unbaked muffins 30 35.82 £ 0 .95 Values are expressed as means - standard deviation. Purity unenriched wheat flour, Maple Leaf Mills, Toronto, Ontario. 95 Table III. Total iron content of flour and bran. Sample Total iron (ppm) flour 10.52 i 0.27 AAGC bran 171.71 1 5-80 # 10 bran 178.88 i 7.12 # 40 bran 190.00 .- 10.39 Values are expressed as mean - standard deviation, n=3« ^Purity unenriched flour, Maple Leaf M i l l s , Toronto, Ont. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096164/manifest

Comment

Related Items