" ' .1 nu- . vwilrlélliiESS T ‘1‘ “31W. OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: ________.______________.___ Place in book return to remove charge from circulation records INFLUENCE OF TWO FORMS OF IRON ON ASCORBIC ACID IN INFANT FOODS BY Catherine E. Adams A THESIS Submitted to ’Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1981 / 3/ 0 //,7 ,5— (30‘5/ ABSTRACT INFLUENCE OF TWO FORMS OF IRON ON ASCORBIC ACID IN INFANT FOODS BY Catherine E. Adams Iron fortification in infant foods is an issue of contention among nutritionists, industry and consumers. Two forms commonly added are ferrous sulfate and electrolytic iron. Ascorbic acid enhances iron absorption, but iron is catalytic in oxidation of reduced ascorbic acid, and may decrease the utility of iron—fortified infant foods as source of vitamin C nutriture. ‘Methodology for analysis of reduced vitamin C and dehydroascorbic acid was evaluated in reference to standards. Observations lead the researcher to question some methods reporting concentration of ascorbic acid in literature. Correction equations were obtained for use with experimental data. . The experiment evaluated mixed cereal (dry cereal with apple juice) with electrolytic iron, and wet cereal with ferrous sulfate, over 10 days of refrigerated storage. Only mixed cereal had significantly greater rate of oxidation with iron present than the non-fortified product. Considering a normal storage period, neither product showed substantially increased reaction rate in presence of iron to warrant concern 0 I DEDICATE THIS WORK TO THE ONES WHO HAVE OFFERED ME SUSTENANCE AND SUPPORT WITH THEIR LOVE THROUGH THE MANY HOURS OF ITS PREPARATION - to my parents, whom I shall always love, to my brother, Chuck, and to a friend, Nancy. 11 ACKNOWLEDGMENTS The author wishes to express sincere gratitude for the guidance and support offered by her graduate committee: to Dr. Jerry N. Cash, major advisor; to Dr. Karen J. Mbrgan, for her moral support and direction offered; to Dr. John L. Gill, for his counsel and success in enabling a student to gain an appreciation of statistics; to Dr. George A. Purvis, for his long-distance guidance and dedication in continuing a commitment to Michigan State university; and a special note of gratitude to Dr. Gilbert A. Leveille for his undaunting advocacy in support for the author in furthering her professional life, for the counsel offered on the project and inspiration for a future career. A most grateful appreciation is extended to Dr. Dennis R. Heldman and Dr. James R. Kirk for their counsel on the project, and for the encouragement offered for both professional and personal growth. The author also expresses appreciation to Gerber Products Company for providing the product used in the project, and a special thanks to Bob Wellace for offering technical assistance. Further, the author wishes to make a sincere gesture in acknowleging the Michigan State University administration and those 1HVO1ved in academic governance who guided the author in personal growth and for knowledge gained and awareness garnered of the fulness of life —- with a special thank you to Provost C. L. Winder. In gratitude to those who have touched the author's life, to 111 her parents and to her friends: Bahia My special thanks, whose evennbalanced soul, From first youth tested up to extreme age, hustness could not make dull, nor passions wild: Who saw life steadily and saw it whole. Matthew Arnold 1822-1888 iv TABLE OF CONTENTS LIST OP TABLES . . . . . . . . . . . .......... . . 1 LIST or FIGURES . . . . . . . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . ....... . . . 1 LITERATURE REVIEW . . . . . . ...... . . . . . . . . . . 4 Dietary Requirements of Infants . . . . . . . ..... . . 4 Prevalence of Iron Deficiency and Anemia . . . . . . . . . 6 Recommendations Made . . . . . . . . . . . . . . . . . . . 7 Reevaluations Offered by Opponents . . . . . . . . . . . . 9 Evaluation of the Prevalence of Breast-Feeding . . . . . . . 10 Effect of Iron Deficiency Exclusive of Anemia . . . . . . . 11 Transitional Foods . . . . . . . . . . . . . . . . . . . . . 15 Iron Bioavailability . . . . . . . . . . . . . . . . .-. . . 16 Recommendation Made For Feeding Dry Cereals to Infants . . . . . . . . . . . . . . . . . . . . . . 17 Iran Enrichment . . . . . . . . . . . . . . . . . . . . . . l7 Enrichment of Dry Infant Cereals . . . . . . . . . . . . . . 19 Further Evidence for the Effect of Small Particle Size and Greater Solubility in Enhancing Bioavailability of.Iron . . . . . . . . . . . . . . . 21 Technical Restrictions in Iron—Fortification . . . . . . . . 23 Influence of Specific Foods on Absorption of Enriched Iron . . . . . . . . . . . . . . . . . . . . 24 Rple of Ascorbic Acid in Enhancing Iron Absorption . . . . . 25 Frospective for Combining Ascorbic Acid with Fortification Iron in Infant Foods . . . . . . . . . . 28 Ascorbic Acid —— Structure and Properties . . . . . . . . .. 30 The Three Forms Identified . . . . . . . . . . . . . . . . . 30 Influence of Acid Conditions on Reactivity . . . . . . . . . 30 Influence with Presence of metal Catalyst . . . . . . . . . 32 Influence on Stability by water Activity . . . . . . . . . . 34 *Nethods of Analysis for Ascorbic Acid . . . . . . . . . . . 37 mmfl PRochRE S O O O O O O O O O O O O O O O O O O O I} 1 Preparation of Model Systems . . . . . . . . .'. . . . . . . 41 Ascorbic Acid Determination . . . . . . . . . . . . . . . . 41 Reduced Ascorbic Acid . . . . . . . . . . . . . . . . . . . 42 Dehydroascorbic Acid . . . . . . . . . . . . . . . . . . . . 42 Diketogulonic Acid . . . . . . . . . . . . . . . . . . . . . 43 Pilot Experiments . . . . . . . . . . . . . . . . . . . . . 44 V EXPERIMENTAL PROCEDURES (cont.) Comparison of Methodology for ReAA . . . . . . . . . . . . . . . 44 Comparison of Methodology for DHA . . . . . . . . . . . . . . . 44 Preparation of Infant Food Samples . . . . . . . . . . . . . . . 45 iron Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 50 Experimental Design . . . . . . . . . . . . . . . . . . . . . . . 52 Pilot Projects for Comparison of Methodology . . . . . . . ... . 53 Ten-Day Storage Study for the Cereal Infant Foods . . . . . . . . 53 Indication of Background Interference with Ascorbic Acid . . . . 54 Iron Evaluation . . . . . . . . . . . . . . . . . . . . . . . . . 54 RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 57 Comparison of Methodology for ReAA . . . . . . . . . . . . . . . 57 Comparison of Methodology for DEA . . . . . . . . . . . . . . . . 63 The Experiment . . . . . . . . . . . . . . . . . . . . . . . . . 67 Total Ascorbic Acid . . . . ................. . . . 67 Reduced Ascorbic Acid. . . . . . . . . . . . . . . . . . . . . . 73 Dehydroascorbic Acid . . . . . . . . . . . . . . . . . . . . . . 75 Diketogulonic Acid . . . . . . . . . . . . . . . . . . . . . . . 83 k-Reaction Rate Constants . . . . . . . . . . . . . . . . . . . . 83 Influence of Oxygen Concentration . . . . . . . . . . . . . . . 93 Influence of water Activity . . . . . . . . . . . . . . . . . . 94 Interaction of water Activity with Metal Catalyst . . . . . . . . 97 Solubility Factor for the Iron Catalyst . . . . . . . . . . . . . 99 Influence of pH on Valence State of Iron . . . . . . . . . . . . 100 Influence of pH on Oxidation Rate . . . . . . . . . . . . . . . 103 Effects of Heat Processing on Iron Form. . . . . . . . . . . . . 104 Iron Content in Samples . . . . . . . . . . . . . . . . . . . . 105 Significance of the Research . . . . . . . . . . . . . . . . . . 106 MY AND CONCIIUSIONS o o o o o o o o o o o o o o o o o o o o o 109 ”PENDICES o o o q o o o o o o o o o o o o o o o o o o o o o o o o 112 I. Splitvplot Analysis —— Total Ascorbic Acid . . ........ 112 II. Split-plot Analysis —— Reduced Ascorbic Acid . . . . . . . . 113 III. Split—plot Analysis -- Dehydroascorbic Acid . . . . . . . . 114 IV. Bonferroni-t Analysis and Student's t Test w— k constants, Dehydroascorbic Acid . . . . . .. 115 V. Split—plot Analysis - Diketogulonic Acid . . . . . . . . . . 116 VI. Bonferroni—t Analysis and Student's t Test —« k constants, Diketogulonic Acid . . . . . . . . 117 'VII} Student's L-test Distribution --Iron Concentration . . . . 118 LIST OF REFERENCES . . . . . . . . . . . . . . . ....... . . 119 V1 6. AI. AII. LIST OF TABLES Page Average nutrient values for baby fOodiproducts per 100 g. .. .46 Correction equations for concentration based on linear regression for each model system.and set of standards. . . 68 krReaction rate constants for mixed and wet cereal products 0 O O O O O O O. 0 O O O O O I O O O O O O O O O O 78 Onenhalf life, in days, for ascorbic acid degradation (1AA.& RrAA) or doubling time (DEA & DKGA); and ratio between non-fortified and iron-fortified product. . . . . 96 Moisture content for infant food products. . . . . . . . . 97 Relative concentrations according to methods of anaIYSis O I O I O O I O O I O O O O O O I O O O O O O O O 109 Split—plot Analysis nnTotal Ascorbic Acid . . . . . . . . 112 Splitsplot Analysis - Reduced Ascorbic Acid . . . . . . 113 AIII. Split-plot Analysis w- Dehydroascorbic Acid . . . . . . 114 ATV. AV. Bonferroni-t Analysis and Student's t Test -- k constants, Dehydroascorbic Acid. . . . . . . . . . .-. . 115 split—plot Analysis —- Diketogulonic Acid . . . . . . . . 116 ARI. Bonferroni-t Analysis and Student's t Test —- k constants, Diketogulonic Acid . . . . . . . . . . . . . 117 AVII. Student's t—test Distribution - Iron Concentration . . 118 vii LIST OF FIGURES Figure ‘ Page 1. Three chemical forms of ascorbic acid produced during ondation O O O O O O I O O O O 9 O O O O D O O O O O C O O 31 2. Catalytic effect for the oxidation of ascorbic acid in the presence of Cu (II) ion at 250 C, as function of pH. . . . . . 33 3. Retention of ascorbic acid as function of aw at 35°C. . . . . 36 4, Schematic diagram.of mixed cereal preparation. . . . . . . . 48 5. Samples prepared for analysis. . . ..... . . . . . . . . 49 6. Split-plot design of the experiment. . . . . . . . . . . . . 57 7. Comparison for RrAA.in 10o Brix model solution with RrAA standards by Autoanalyzer and titrametric methods. . . . . . 58 8. Comparison for RrAA in 5% rice starch model solution with RPAA standards by Autoanalyzer and titrametric methods. . . . 60 9. Formation of a fluorescent compound. . . . . . . . . . . . . 61 10. Borate-carbohydrate complexes. . . . . . . . . . . . . . . . -62 11. Comparison for DHA in 10° Brix model solution with DHA standards by Autoanalyzer and spectrophotometric methods. . . 64 12. Comparison for DHA in 5% rice starch model Solution with DHA standards by Autoanalyzer and spectrophotometric methods .66 13. Change in TAA concentration ratio for non-fortified and ironwfortified cereal products during 10 days of refri- Se-rat ed storage I O O O C O I I O O O O O O O O C O O O O O 0 7o 14. Ratio for difference in ascorbic acid metabolites for .mxe.d c e re a 1 C O C O O C C O C O O O O O C C C O O C O O O O 7 1 15. Ratio for difference in ascorbic acid metabolites for wet cereal O O O O O C O I O O O O O O D O O O O O C O O O O 72 16. Change in RsAA concentration ratio for non-fortified and ironrfortified cereal products during 10 days of refri- gerated storage. . . . . . . . . . . . . . . . . . . . . . . 74 viii Figure Page 17. The influence of iron on DEA in mixed cereal and wet cereal O O O O O O C C O C O O O O O O O O O O O O O O O O O 7 7 18. Change in DEA concentration ratio for non-fortified and iron-fortified cereal products during 10 days of refri- gerated storage 0 I O O O O O I O O O O O O O O O O O O O O O 79 19. Change in DEA concentration ratio for non-fortified and ironnfortified mixed cereal. . . . . . . . . . . . . . . . . 81 20. Change in DEA concentration ratio for non-fortified and iron-fortified wet cereal. . . . . . . . . . . . . . . . . . 82 21. k.First-order reaction rate for ascorbic acid metabolites in mixed cereal. . . . . . . . . . . . . . . . . . . . . . . 85 22. k First-order reaction rate for ascorbic acid metabolites in wet cereal O O O O O O O O 0 O O I O O O O O O l O O I O O 87 23. k—Rate constants for TAA in mixed and wet cereal products . .89 24. krRate constants for RrAA in mixed and wet cereal products . 90 25. krRate constants for DEA in mixed and wet cereal products. . 91 26. Adsorption isotherm for the dehydrated model food sysytem 0 at 20 C. . 0 Q Q C Q C C Q C I O O C O C O O C O I O O C O O 98 27. Semilog plot of pH vs. the percentage of added iron converted to ferrous ion after 48 hrs. . . . . . . . . . . .101 ix INTRODUCTION Nutrient fortification of several foods is considered commonplace by consumers today, and it is intended to provide an adequate supply of nutrients that may be consumed at low levels in the diet of some individuals. The fortification of table salt with iodine has been credited with eliminating the widespread prevalence of goiter in parts of the United States. The state of Michigan, in 1921, reported 47 percent of school children as having goiter. The use of iodized salt was extensively promoted, and in 1951, only 1 percent of the population was reported to have simple goiter (Gutherie, 1975). Iron is an essential nutrient that has been accepted in fortifica- tion of white.bread and flour since 1941, as result of an executive order enacted during Waldhhr II. The effort in fortification of foods with iron was intended to combat the incidence of iron deficiency anemia. Limited diets, such as those consumed by infants, may not include all essential nutrients and in the quantities required for optimal growth and physiological development. Some foods that traditionally comprise a major portion of an infant diet have been fortified and enriched to contain significant amounts of many needed nutrients we iron and vitamin C are.two nutrients that are commonly added. 4 Two forms of iron are currently added to dry-infant cereal and wet infant cereal packed in jars. The two forms, electrolytic iron and ferrous sulfate respectively, differ in their bioavailability with ferrous sulfate as the more readily absorbable form. The two forms also differ in respect to their potential food applications. Reduced ferrous sulfate is not 2 appropriate for use in dry cereal since the particles precipitate from an even distribution in a package; and with.time, the cereal becomes grey in appearance from a chemical reaction with the reduced iron. Presence of vitamin C with iron has been shown to enhance iron absorption (Bjorn-Rasmussen and Hallberg, 1974), yet iron is also recog— nized as a catalytic agent in promoting oxidation reactions that degrade the quality of reduced forms of nutrients. The objective in fortification is to promote health, yet if the presence of iron with vitamin C leads to an increased rate of degradation for the latter nutrient relative to when iron is not fortified, then its addition in baby foods would not be advised. Therefore, a project was initiated to evaluate the influence of two forms of iron on ascorbic acid in two products where both nutrients are supplied in substantial amounts. No research had previously demon- strated change over time for concentration of various forms of ascorbic acid in the opened product. Vitamin C is declared on the nutrient label as the amount of reduced vitamin present after a specified period of shelf storage. Yet, nutrient content of foods as consumed may be substantially different in product once opened and stored for a period of time. Also, the interaction with other food components may be.active in altering nutrient composition. The nutrient profile of baby foods, as consumed, was evaluated to reflect the-influence.with presence of the two forms of iron in their respectivexproducts. It was of primary interest to establish the rate of change for ascorbic acid contrasting the two products in respect to iron presence or its absence. By evaluating the reaction rate as influenced by iron, the utility 6f ironefortification and vitamineC enrichment of baby foods could be established and reported. LITERATURE REVIEW Dietary Requirement g§_lnfants Individuals are more vulnerable to nutritional inadequacies during particular stages of the life cycle. The Surgeon General reports that the health of the American people has never been better (U.S.H.E.W., 1979). The present situation is attributable in part to advances in technology which have provided Americans with a nutritious, consistent and available food supply. However, nutrient needs are high during periods of physical growth and development, or emotional stress. During these times when change is occurring, the demand that the diet supply required amounts of all nutrients is challenged, and nutritional supplements sometimes play a role in meeting needs. Some nutritionists advocate that nutrient requirements be fulfilled by including a wide variety of foods from the four basic food catagories without reliance on the use of vitamin/mineral supp- lements (N.A.S., 1973). Therefore, when nutrient needs are high, more careful consideration should be given to the diet to ensure that requirements are met. Infancy is a time when development of all physiological systems are in a critical state and the nutritional needs of the infant are of particular concern. Further, some researchers suggest that certain adult characteristics, including mental retardation, obesity, hypertension, and atherosclerosis have their origins in early feeding patterns (Dairy Council, 1979). 4 5 Nutritional research, as directed by the recommendation of the Food and Nutrition Board of the National Research Council (Food and Nutrition Board, N.R.C., 1979) has focused on prevention of degener- ative diseases that may be incurred naturally during the aging process. Many of the major health problems in the U.S. are known to be of multiple etiology. Attainment of optimal nutrition during infancy with its continuation throughout the life cycle may contribute to a reduced incidence, or delayed onset, of some of these illnesses. The controversy regarding breast-feeding has been reviewed (Jelliffe, 1975; Gerrand, 1974; Oseid, 1975, Fomon, 1974c). It is generally agreed that a distict advantage of breast-feeding is that human milk is best suited to the infant's nutrient needs during the earliest stage of growth. Additional advantages include promotion of an harmonious mother-child relationship and, in some situations, the economic advantage of breast-feeding relative to artificial feeding. Ousted and Sleigh (1975) and the American Academy of Pediatrics (1976) have recently recommended that infants be breast-fed for the first four to six months of life. It is acknowledged, however, that there is a certain segment of the population that cannot breast-feed for anatomic, physiologic, or toxicologic reasons. While breast-feeding may provide protein, calories, most minerals, vitamins, and fluids in proportions best suited to infants' needs, it is not intended to provide sufficient nourishment for growth during the entire first year or two of life. Human milk does not contain sufficient amounts of iron to meet infants' requirement (Brown, 1973). The concentration of iron in breast milk is only slightly higher than that of cow milk or unfortified cow milk formula (Underwood, 1971; 6 Murthy, 1971), yet iron in breast milk is uniquely bioavailable (McMillan, 1976; Saarinen, 1977). The need for supplemental iron in infants on prolonged breast-feeding is, therefore, likely to be less critical than in those fed unfortified cow milk formula. The newborn infant is well supplied with iron, both in terms of iron needed for metabolic and enzymatic functions. Eighty percent of iron stored is in hemoglobin (Sjolin, 1968; Smith, 1974) and the remainder is in such storage forms as ferritin and hemosiderin which function in iron homeo- stasis. Iron requirements of pregnancy are high (Recommended Dietary Allowances, 1974), yet iron deficiency at this stage has greater conse- quence on the developing fetus, since nutrients available in only limited amounts are preferentially transferred to the fetus (Widdowson, 1951). Within the first year of life the infant must double its body iron as it triples its body weight, and transplacental delivery of iron that guaranteed iron nutrition for the fetus is replaced with a smaller and less consistent dietary supply (Dallman, 1980). Iron stores are depleted during early development, but it is not until the fourth month of life for the term infant that the increased dependence on dietary iron is apparent. Prevalence g§_Iron Deficiency and Anemia Iron deficiency anemia (hypochromic and microcytic) in infancy has been defined as the state in which hemoglobin concentration is less than 11 mg/dl and that which responds to the administration of iron (Fomon, 1974a). Iron deficiency anemia has been recognized as the most prevalent nutritional disorder among infants and children in the U.S., particularly those between the ages of six and 24 months (Fomon, 1970). A low iron content of either breast milk or unsupplemented cow milk formula (about 0.5 mg/liter) (Fomon, 1974b) predisposes the infant to depleted iron reserves. The total iron store in preterm.and low~birth~weight babies is lees than normal at birth and a rapid rate of postnatal growth quickly exhausts reserves before four months. Lundstrom et a1. (1977) observed iron deficiency in lowbbirthdweight infants between the ages of two and three months. Thus, it is clear that an exogenous source of iron is even more critical for the preterm than for term.infants (Lundstrom, 1974; Schulman, 1954; Committee on Nutrition, American Academy of Pediatrics, 1976). Filer (1969) reviewed the significance of iron—deficiency anemia as a public health problem and concluded that its existence is less pre- valent in more favorable socioeconomic circumstances and in large metropolitan centers (Haughton, 1963; Andelman, 1966; Danneker, 1966). Haddy and coeworkers (1974) evaluated iron status in 109‘infants and children ages four months to two years from low-income families. Incidence of iron deficiency (hemoglobin concentration less than 11 mg/dl and hematocrit values lower than 332) was slightly more than 50 percent. While incidence of iron deficiency is lower in more affluent socio- economic strati, Sturgeon (1959) found approximately seven percent of term infants, not consisdered underpriviliged, with hemoglobin levels lesstflun110 mg/dl by one year of age. Further, Hutcheson (1968) reported that 10 percent of Tennessee‘s rural poor children under six years of age were anemic. Recommendations Made The high risk of iron deficiency, with or without clinical 8 symptoms of anemia, prompted the Committee on Nutrition of the American Academy of Pediatrics to issue several recommendations over recent years for iron supplementation of infants and children (Committee on Nutrition, 1969; Committee on Nutrition, 1971; Committee on Nutrition, 1976). The first recommendation was made in 1969 when the Committee published a review of iron requirements for infants. The Committee urged that pediatricians recommend inclusion of iron-fortified foods in the infant's diet for at least 18 months. In 1971, a statement was issued recommending the use of ironnfortified formula until at least 12 months of age. Additionally, it was suggested that iron-fortified whole milk or iron-fortified evaporated milk be made available for infant feeding. Ironesupplemenation in breast-fed infants was not specifically advised due to the enhanced bioavailability of iron in r breast milk. The Committee faced both praise and demonishment by numerous professionals for thses recommendations (Schafer et al., 1971; Haddock and Harvey, 1971; Diamond, 1971; Melhorn, 1971; Filer, 1971a; Fisher and Allen, 1971; Filer, 1971b; Pearson, 1971; Oski, 1972; Pearson, 1972; Lopez, 1972; Feurth, 1972). Opponents argued that infants from middle-class families were not as likely to incur iron-deficiency anemia as would those from lowwincome groups (Owen et al., 1970; Burks et al., 1976) and would thus not benefit from iron supplementa- tion. Kripke and Sanders (1970) reported prevalence of iron—deficiency anemia among children ages six months to three years seen at ambulatory clinics in Iowa. Anemia was identified in only 4.1 percent of a sample whose socioeconomic status was different than previous groups that had been studied in low—income areas of large metropolitan centers. 9 It is argued that iron additives given to infants with an already adequate iron status could result in an iron overload (Danneker, 1966). Infants with hemolytic diseases such as sickle cell anemia or a form of hereditary hemochromatosis would be particularly susceptible to dangers of excessive iron intake. In light of the rare occurence of such diseases, ‘more concern was voiced with regard to the ultimate outcome of eating additional iron-fortified cereal and possibly iron-fortified bread over an extended period of time. Some physicians continued to be concerned with possible increased incidence of feeding problems or gastrointestinal disturbances although no documented evidence has been demonstrated that this was a significant problem (Committee on Nutrition, 1971). VReevaluations Offered bz_gpponents Owen et a1. (1970) reported that anemia was not as prevalent in American preschool children as might have been anticipated from earlier reports. However, Owen and co-workers (1971) reported that, while anemia predominated among preschool children from.lower income families, iron- . deficiency (based on levels of plasma iron) was relatively common in all segments of the population. In a recent communication from Ahn and Maclean (1980), data was presented indicating that growth of infants receiving human milk as a sole source of protein and energy was comparable to that of the National Center for Health Statistics (NCHS) population up through the ninth month of life. Consequently, it was implied that human milk alone provides adequate nutrition for growth of an infant during this period of time. In closer comparison of the NCHS growth - curves, it was demonstrated that the curve tended to successively cross lower percentile lines towards the end of the 10—month observation period. 10 This was particularly true for male children. This observation denotes a declining ability of human milk to adequately serve as sole source of nutrients for the infant towards the end of this 10~month period (Morley, 1976). Values did, nonetheless, stabilize between the 25th and 75th percentiles before the end of 10 months and did not indicate imminent malnutrition. Evaluation gf_the Prevalence gf_Breast-Feeding The nutritional and psychological merits of breast-feeding warrant its continued support as a method of feeding the newborn infant. Prevalence of breast-feeding in the U.S. has fallen 100 percent over the past 75 years and less than one percent of one- year-old infants in previous years were breast-fed (Bain, 1948; Hill, 1967). Recently, there has been renewed interest in breast- feeding, particularly among more educated mothers and the declining trend is reversing (Jelliffe, 1976). Reports have surfaced that morbidity and mortality rates for breast-fed infants are generally lower than for artificially-fed infants in underdeveloped countries (Mata et al., 1971; Mellader, 1959; Adebonojo, 1972; Jackson et al., 1974). Similar rates have been reported comparing infants in developed countries. Reasons for lower rates of mortality and morbidity existing for breast-fed infants appear complex, but a major contribu- ting factor appears that supplemental solids or cow milk and formula derived thereof may easily become contaminated since bottles and nipples may not be adequately sterilized. Further, water supplies may not be sanitary, and may thus bottle—feeding may become a vehicle for infection and disease. A newborn infant has not developed the 11 capacity to combat even lowwlevel insult from baCterial contamination that predominates in some underdeveloped communities and an increase in morbidity and mortality rates of infants results (Mata et al., 1971; Mbnckeberg, 1973; Plank and Milanesi, 1973). Economic implications of breast-feeding are salient in developing countries and in areas of the U.S. where a choice to bottle-feed an infant may have serious economic consequences. The purchase of formula and manufactured weaning foods may divert scarce monetary resources for families in developing countries who earn an average of 4 U.S. cents per day (Latham, 1979). Such purchases may jeopardize the health and nutritional status of other family members. Effggt_gf_lron Deficiency Exclusive 2f_éngmiat' Among other advantages, human milk gives the infant immunologic protection against respiratory and gastrointestinal infection as well as allergic reactions (Dairy Council, 1979). While some of the cellular components considered beneficial in human milk are not secreted after 5 days post partum (Lawrence, 1980), there are both cellular and humoral factors associated with breast milk that may have long—term advatages for the infant. Cellular lymphocytes synthe- size the antibody IgA, which is the most important immunoglobin in milk -- not only in concentration, but also in biological activity. The IgA content of human milk is very high in colostrum.and drops precipitously after 4 to 6 days, but remains constant thereafter for as long as 27 months. The "bifidus factor" in human milk facilitates the growth of Lactobacillus bifidus which appears to have an 'intestinal guardian' 12 function in checking the proliferation of undesirable and possibly harmful organisms, such as pathogenic E, coli (Jelliffe and Jelliffe, 1978). Colonization of the alimentary canal for breast-fed infants is primarily with L, bifidus, whereas the intestinal flora of babies fed cow milk is primarily gramrnegative bacteria, especially coliforms and ' bacterioides. However, prolonged duration of breast-feeding does little to replete exhausting iron stores, and if deficiencies are allowed to persist will result in nutritional anemia. Other symptoms of iron- deficiency can exist without clinical signs of anemia and may easily go undiagnosed and untreated. While it is difficult to separate the physiological consequences of deficiencies in tissue iron from those of anemia, there has recently been some success in making this distinction. This has been by the demonstration of impaired exercise tolerance in the rat and defects with the immune response in man. Both conditions may occur independent of anemia (Dallman et al., 1978). The symptoms of iron—deficiency are due partly to the comprimised delivery of oxygen to tissues resulting from a decrease in concentration of hemoglobin (Anderson and Barkue, 1970; Viteri and Torun, 1974; Gardener et al., 1975). Iron-deficiency may also result from depletion of iron-containing compounds in tissues (Jacobs, 1969; Fairbanks et al., 1971; Dallman, 1974), and may thus give rise to clinical mani- festations of anemia. Tissue iron may be classified in several catagories. The first, heme iron compounds, exist in cytochromes and myoglobin and are respon- sible for the oxidative production of cellular energy in the form of adenosine triphosphate (ATP). Myoglobin is essential for storage of 13 oxygen and functions during muscle contraction. Second, the non—heme related iron-containing compounds are the iron-sulfur proteins and metalloflavoproteins. These compounds account for more iron in the mitochondria than with the cytochromes, and include the enzymes NADH dehydrogenase and succinic dehydrogenase. These are essential in the functioning of cellular respiration. A third classification of compounds do not .eontain iron but rather require iron as a cofactor. Iron- deficiency that has progressed beyond simple depletion of iron stores will affect the tissue iron compounds in a non-symmetric manner that? cannot fully be predicted from.the degree of anemia (Dallman, 1974). Biochemical and morphological abnormalities result from depleted iron reserves and may be related to aberrant physiological functioning. Researchers attribute the anemia of iron-deficiency in striated muscle for an observed decreased work capacity (Anderson and Barkue, 1970; Viteri and Torun, 1974; Gardener et al., 1975). Finch (1976) showed in rats that striated muscle dysfunction is not imediately reversed by transfusion to correct the anemia, but may be alleviated with four days of iron therapy. Iron-deficiency has been implicated in adult patients with altered behavior including apathy, irritability, and inability to concentrate (Dallman et al., 1978).Da11man (1975) reported that ironedeficiency initiated early in life for the rat results in deficient cerebral iron which persists even after iron repletion of other tissues. Mackler et a1. (1978) corroborated earlier reports of a casual relationship with iron-deficiency in his observa- tion of decreased activity for aldehyde oxidase, a key enzyme in the pathway of serotonin degradation, and observed an elevated concentration of serotonin in brain tissue of iron-deficient rats. 14 Further abnormalities resulting from ironndeficiency that involve the central nervous system.are associated with elevated urine catechol- amine levels. webb and Oski (1974) proposed that elevated levels were the consequnce of the decreased activity for monoamine oxidase, an enzyme involved in the catabolism of neural mediators. In addition to mental changes, alteration of catecholamine concentration can explain the increased irritability and abnormalities in appetite sometimes described in association with iron-deficiency. A condition frequently associated with iron-deficiency is an increased susceptibility to infection (Pearson and Robinson, 1976). The infant, with its yet undeveloped immune system is particularly vulner- able to disease. Several research teams have provided evidence of impaired lymphocyte and neutrophil function as a basis for such a relationship (Arbeter et al., 1971; Johnson et al., 1972; Chandra, 1973; thdougall et al., 1975; Srikantia et al., 1976). The lympho- cyte abnormalities may be related to a lower incidence of positive skin tests for iron-deficient than for iron-replete groups when tested with a purified protein derivative and Candida antigen (Joynson et al., 1972; Macdougallet al., 1975). Neutrophils are defective in the oxidation-reduction of a nitro blue tetrazolium dye, suggesting that the iron-containing enzyme for this reaction is present in diminished amounts. Gastrointestinal functioning may also be impaired when iron re- serves are depleted. Ghosh et a1. (1972) reported a reduction in acid secretion by the stomach with iron-deficiency. Other researchers have found that intestinal absorption (Yeoman and St. John, 1975) including iron absorption (Kimber and weintraub, 1968) is impaired without an 15 adequate supply of tissue iron. Biochemical deficiencies of the intest- inal mucosa of the rat that affect cytochrome enzymes were thought to be the cause for decreased absorption. Thus, knowledge that iron stores are depleted by four to six months of age in the term infant suggest that some type of iron supplementation should be considered. Saarinen (1978) found that iron supplemented infants rarely had laboratory evidence of exhausted iron stores. Clinical symptoms of anemia are not always apparent with sub- optimal iron status, yet the long-term consequence of inadequate iron nutrition has been documented. Transitional Foods In view of the 1976 recommendation by the Committee on Nutrition of the American Academy of Pediatrics (1976) for iron fortification of infant foods, Saarinen (1978) evaluated iron status of infants on prolonged breast-feeding relative to others receiving a cow milk formula and infants receiving a proprietary infant formula. Results indicated that supplemental iron should be considered for breast-fed infants after six months of age. In contrast, infants who were weaned to cow milk needed iron supplementation earlier -- even at four months of age or before. It was speculated that increased bioavailability of breast milk accounted for the observed enhanced iron status for breast-fed infants relative to those receiving the cow milk formula. Further, Siimes et a1. (1979) reported that iron concentration in breast milk declines during the course of lactation and infants on prolonged breastsfeeeding are forced to depend on a depleting iron reserve. Transitional solid foods may be readily introduced to the l6 infant's diet at four months of age and may supplement breast- or bottle- feeding (Cook et al., 1972). Saarinen (1978) reported that non-fortified solid foods do not provide a sufficient source of iron for infants on prolonged breast-feeding. It was suggested that among solid foods con- sumed by infants, iron-fortified cereals could provide a major source of iron and would be expected to prevent iron-deficiency during the latter portion of infancy. Some nutritionists have attempted to discourage dependence on vitamin/mineral supplements and instead suggest that individuals select a well-balanced diet to attain optimal nutritional status( N.A.S., 1973). It is important to consider dietary recommendations aimed at improving iron nutrition from.an economic as well as nutritional perspective, and additionally to develop habits that will be beneficial in later life. Opportunity is limited for an infant to consume a wide variety of foods, since the ability to digest some foods or physically handle foods is not yet developed. In the U.S., the inclusion of iron-fortified dry cereal with introduction of solid foods in the infant's diet is a convenient and practical mode to attain adequate iron nutrition. Iron Bioavailability It is recognized that the availability of iron from foods depends greatly on several factors including body iron stores, form and amount of iron in foods, and the combination of foods in the diet (Monsen et L al., 1978). Cook and co—workers (1972) in their report implied that the unusually high bioavailbility of breaatmilk-iron is modified by the introduction of various solid foods. The inclusion of meat in the diet has been found to profoundly enhance absorption of all forms of food 17 iron (Layrisse and Martinez—Torres, 1971). Further, the form of iron in meat that exists in heme protein is more readily absorbed than is non- heme iron in vegetable and grain products. In one study, using human subjects, 37 percent of the heme iron was absorbed from.a test meal in contrast to only five percent of the non-heme iron (Bjorn-Rasmussen et al., 1974). There is almost no heme iron in the infant's diet when milk is the major source of calories. The relatively small amount of iron present exists primarily in the non-heme form. Emphasis on meat in an infant's diet may facilitate formation of adequate iron reserves, but such focus may not be economically practical and may be excessive in respect to the infant's capacity to digest high protein foods. Cereal products may be effectively utilized in preventing iron-deficiency within practical constraints, providing that such foods are iron- fortified at adequate levels with a form of iron that is both available and the application technique technologically feasable. Recommendation Made For Feeding Dry Cereals £g_lnfants The Committee on Nutrition of the American Academy of Pediatrics (1976) made the recommendation that current nutrition counseling should emphasize the use of dry infant cereals to provide an adequate source of iron in the infant's diet. The recommendation included that dry cereals be used during the first and second year of life since requirements for iron continue to be large during this extended period. Iron Enrichment The concept of iron enrichment for cereals has found forum.for contention with legislators, nutritionists, and with the milling and 18 baking industry. The alarming prevalence of low iron intakes reported in the Ten-State Nutrition Survey (1971) led the FDA and several nutritionists to propose a 300 percent increase in the present mandate for the amount of iron required to be added to enriched white bread in the United States (Abraham et al., 1974; Finch and Monsen, 1972; Goldsmith, 1973). Inadequacies of this report have been widely discussed (Crosby, 1974). However, the enrichment of flour and bread with vitamins and iron has been practiced since 1941, when it was initiated as one of the measures taken to handle the food emergency encountered as result of World War II. The levels of enrichment currently in practice have been in effect since 1943. However, evidence that persistence relating to the prevalence and severity of iron-deficiency anemia even with the current iron enrichment program of cereal products lends justification to doubt the efficacy of such a practice. Nutrition surveys of adult U.S. Army personnel in basic training from the period 1970 to 1971 revealed surprisingly high prevalence of anemia among both young males and females (Col, 1972). The infant and young child require relatively greater amounts of dietary iron in order to maintain stores. When iron-deficiency anemia is observed in adolescents and adults, it is generally accepted that this is a reflection of iron stores never having attained adequate levels in infancy (Council on Foods and Nutrition, 1972); and thus, the patent importance of establishing adequate iron nutrition in the infant is indicated. Medical nutritionists, pediatric and hematological groups concerned with iron-deficiency anemia have been expressing their views for many years, yet the controversy of cereal enrichment with 19 iron intended for infant feeding continues to provide forum for debate. Primary contention concerns selection of the particular source of iron used in enrichment of cereal-based foods. Stipulation of the proposal in the Federal Register (1971) is that added iron be in a form which is harmless and assimilable. All chemical forms of iron used in foods must meet the criterion of safety, as do all food additives, but exper- imental animal data indicate that bioavailability varies widely depending on (a) the specific chemical form of the iron, (b) the food to which the iron is added, (c) the overall diet of which the enriched food is a part, (d) the particular type of processing for the food product, and (e) nutritional status of the individual. Research results of bioavailability in animals is abundant with respect to most forms of iron currently added to foods, yet there is a paucity of data regarding availability of iron from food in the human (Council on Foods and Nutrition, 1972). Enrichment gf_Dry Infant Cereals Most of the dry infant cereals produced in the late 1940's and 1950's contained sodium iron pyrophosphate or other sources of iron which were characterized with probably less than one percent of the iron actually being absorbed (Committee on Nutrition, 1976). A report by Rios et a1. (1975) discredited the value of iron-fortified cereals in infants' diet claiming that sodium iron pyrophosphate and ferric orthophosphate were so poorly absorbed from.infant cereal that iron supplemented cereals failed to provide an effective or predicatable source of dietary iron. Inzicomment regarding the statements made by Rios and associates (1975), Stewart (1975) maintained that Gerber 20 Products initiated substitution of reduced iron for sodium iron pyro- phosphate in its cereals in 1970. Specification of iron particle size requires that 95 percent of the iron passes through a 325 NBS mesh screen, or particles must be 44 microns or less in diameter. Fritz et a1. (1975) reported that bioavailability of such iron was estimated to be 34 percent that of ferrous sulfate (as standard) compared with 14 percent for sodium iron pyrophosphate. In 1972, electrolytic iron was introduced as the form for fortification in cereals manufactured in the U.S. and remains as the form currently in use. Specifications include that 80 percent of iron is less than 20 microns in particle size. In addition, it is reported that a higher solubility index for iron is indicative of increased bioavailability (Rios et al., 1975: Waddell, 1974), and electrolytic iron with a large porous surface enhances solubility relative to other forms of reduced iron powder. In 1970, electrolytic iron was selected for use in cereal products because it is finely divided in cereal with the largest surface area of all iron products available at the time. It is still in use since no compound with greater bioavailability that is equally appropriate for use in cereals has been found. Anderson et a1. (1974) experimented using miniature pigs and compared bioavailability of various forms of iron in rapidly growing animals. Experimental formfiof iron were compared in a cerealdmilk mixture similar to that commonly fed to infants. It was determined that the effect of iron-deficiency anemia on growth performance of animals was statistically significant by the fourth week of the study. At that time, gain in body weight was highly correlated with 21 concentration of hemoglobin in blood. Iron status of pigs fed a diet fortified with sodium iron pyrophosphate did not differ statistically from that of pigs whose diet was not iron supplemented. However, elctrolytic iron of small particle size was 70 percent as available as the reference iron source, ferrous sulfate. Therefore, efficacy of supplementing deficient diets with sodium iron pyrophosphate is questionable, but research results supported supplementation with electrolytic iron of small partfiie size. Further Evidence for the Effect QEDSmall Particle Size and Greater Solubility‘in.Enhancing_BioavailabilityIgf_Iron Shah and Belonje (1973) had claimed that bioavailability of iron. is dependent on particle size. A research team of Fritz and Pla, along with numerous codworkers, have examined bioavailability of several forms of fortification iron (Fritz et al., 1974; Pla and Fritz, 1971). Pla and Fritz (1971) used the hemoglobin repletion method for measuring bioavailability whereby anemic animals respond to supplemental iron feeding and clinical symptoms of anemia are no longer present. Researchers consistently found that electrolytically reduced iron of small particle size was more assimilable than were other forms of iron previously used for enrichment purposes. Electrolytically reduced iron 7 to 10 microns in particle diameter had relative biological value (RBV) of 63.5 1110.9 when ferrous sulfate (Fe804'7H20) was used as the reference standard (FeSO4-100). Other forms of iron were less available. Similarly prepared electrolytic iron of larger size (27 to 40 microns) was also less available with a RBV of only 37.9 1112.2. It has been established that bioavailability of elemental iron 22 powders depend to a large extent on particle size distribution (Hoglund and Reizenstein, 1969; Pla et al., 1973; Motzok et al., 1975), and also on the manufacturing process (Pennell et al., 1975; Pla et al., 1976). The method of preparation for iron powders influences solubility and porosity of surface area. Pla et al.( 1976) and Motzok et a1. (1978) demonstrated that solubility rate shows good promise as predictor of the relative bioavailability for some types‘of elemental iron powders. Motzok et a1. (1978) compared several methods for manufacture of food grade elemental iron powders in regard to particle size, RBV and solubility. Electrolytic and carbonyl iron powders were well utilized by rats, while iron powders produced by reduction at high temperatures with gases such as CO and H2 were poorly utilized by rats. Respective RBV (mean of 12 to 15) for these samples was only one-third or less of the value obtained with feeding electrolytic iron. Particle size may account for some variability in RBV between types of iron powders, and the high proportion of very fine particles (69 percent less than 7 microns in diameter) in a carbonyl iron sample may have accounted for its relatively high solubility in 0.2 percent HCl. Electrolytic iron samples contained particles about 37 percent of which were less than 27 microns in size, while iron powders produced by reduction with H2 and C0 gases contained high proportion of particles whose size was greater than 19 microns. Pennell and co- workers (1975) also reported that the high proportion of annealed surface structure of elemental iron produced by reduction at high temperatures contributed to the low solubility and boiavailability of this type of iron powder. When using this discrete fraction of various iron powders, all of similar particle size, effect on RBV 23 could be assessed for the different methods of manufacture. Small particle size fraction (7 to 10 microns) of electrolytic iron had RBV only 5 to 10 percent higher than those comparable fractions of H2 and CO reduced iron powders, whereas RBV and gaugitgg.solubility of this fraction of electrolytic iron was two times that of the H2 and CO reduced iron. Thus, fortification with electrolytic iron appears to be the most efficacious mode for prevention of iron-deficiency via iron-fortifi- cation programs with iron powders. An imprt‘sn‘t consideration remains that research reported herein are in gi££g_studies. Application to human infant feeding situations is speculative - or presumptive at best. Animal research with depletion- repletion feeding trials is not a realistic physiologic iron status, but conclusions may serve as indication for human iron nutriture. Technological Restrictions in_lron-Fortification _ Iron-fortification and enrichment of food products in the United States has focused on the use of four iron compounds. These compounds include ferrous sulfate, powdered elemental iron, and two insoluble iron salts: ferric orthophosphate and sodium ferric pyrophosphate. Other iron compounds account for less than 5 percent of the total number of forms used (Waddell, 1974). The technical problems encountered with iron enrichment are greater than the difficulties overcome for the addition of vitamins. The soluble forms of iron undergo notable changes in cereal products during storage including the development of rancidity with changes in color, odor,and baking performance. Thus, insoluble phosphate salts were commonly added to cereal foods (Council on Foods and Nutrition, 1941) even though biological efficacy of such 24 practice was questionable (Rios et al., 1975; Fritz et al., 1975). Reduced iron reportedly imparts a grayish appearance in white or lightly colored foods. The two insoluble phosphate salts are light colored which makes their addition particularly applicable to cereal products. While it is accepted that ferrous sulfate is the most bioavailable form from the standpoint of iron nutrition (Pla and Fritz, 1971; Amine et al., 1972; Fritz and Fla, 1972; Ranhotra et al., 1971), use of this form in cereal products presents problems that cannot be tolerated in foods that face extended shelf-storage. The soluble ferrous salt can be used in a strained food system.that is heat processed and is thus more shelf-stable. Recommendation of the Committee on Nutrition of the American Academy of Pediatrics (1976) includes that ready-to-feed cereals with fruit in jars be consistent in iron content and form, and that the form supplied be well-absorbed. Waddell (1974) reported the superior bioavailability of ferrous sulfate as compared to less soluble sources whether the iron was added directly to the diet or was consumed in the form.of enriched bread. Fritz et a1. (1970) also determined that ferrous iron, whether it be in the chloride form or sulfate salt, was some- what better utilized in both rats and chicks than were comparable ferric salts. Influence gf_Specific Foods 22 Absorption of Enriched Iron It has already been noted that the amount of iron available to improve iron nutrition via fortification depends on factors including form and level of iron-fortification, iron status of the individual, and composition of the diet with foods consumed concurrent with supple- 25 mental iron CMonsen et al., 1978). It is well recognized that the critical relationship of dietary iron to establish iron nutrition is accentuated for the infant relative to the adult. Specifically, about 95 percent of the iron required for the production of red blood cells in the adult is recycled from the breakdown of senescent red cells and only 5 percent is derived from dietary sources (Hillman and Finch, 1974). It is estimated that the one-year-old infant derives less than 70 percent of red blood cell iron from.senescent red cells and therefore requires about 30 percent of iron needed for the production of red blood cells from the diet. In order to avoid the need for costly prophylactic supplementa- tion with iron, it becomes increasingly desirable to enhance iron absorption from foods consumed. The relevance of type of fortification iron used and bioavailability has already been discussed. The role of other dietary components in influencing absorption is worthy of further consideration. Role 2f_Ascorbic Acid i2_Enhancing_Iron Absorption The enhancing affect with the inclusion of meat in the diet, referred to as the "meat-factor" (Cook and Mbnsen, 1976), has been mentioned. That presence of ascorbic acid has an enhancing influence has been widely reported (Apte and Venkatachalam, 1965); Conrad and Schade, 1968; Herbert et al., 1966; Layrisse et al., 1968) and is particularly beneficial in respect to absorption of the less available non-heme iron compounds (Cook and Monsen, 1977). Non-heme iron in foods exists primarily in an inorganic ferric (III) iron complex. These complexes are broken down during digestion and the iron is partly reduced to the more readily absorbed ferrous (II) iron form. This conversion process is facilitated by the presence of endogenous 26 hydrochloric acid in gastric secretions and by ascorbic acid (Dallman, 1980). That the conversion takes place prior to digestion in the stomach has been reported. Hodson (1970) investigated a liquid weight-control dietary fortified with iron as ferrous sulfate or ferric orthophosphate. The beverage contained an excess of ascorbic acid. After two to five months storage,the iron added in the trivalent ferric form had been solubilized, ionized and reduced to the bivalent form while the ferrous salt remained in its original soluble form. Lee and Clydsdale (1979) found 85 percent conversion from.aqueous ferric chloride to the ferrous ion in canned beans and a rehydrated breakfast beverage. Lee and Clydsdale (1980) further highlighted the potential for changes to occur in iron valence during processing and storage with an investigation studying the effects of various forms of iron in an ascorbate-fortified beverage over three days of storage. Iron, as ferrous sulfate, remained 100 percent soluble but a slight increase was observed for amount of soluble complexed iron. When iron was added to the beverage containing ascorbic acid as ferric orthophosphate, 21 percent of the iron became solubilized with 17 percent being converted to the more biologically available ferrous form. Iron added in the elemental form became solubilized rapidly and no iron remained as elemental iron after four days of storage. The loss of elemental iron occured geometrically with 90 percent dis— appearing the first day after hydration. After four days, the iron profile of the beverage containing ascorbate and elemental iron was practically identical to that with ferrous iron added. In a later report, Nojeim and Clydsdale (1981) reported that at pH 4.2, over 90 percent of elemental iron present in a phthalate/HCl/ NaOH buffer system with ascorbic acid was ionized to the ferrous form 27 regardless if ascorbic acid was added or not. Gradual oxidation to the trivalent state was observed with time. After four weeks, approximately 25 percent of the ionic iron was present in the ferric state. Physiological amounts of ascorbic acid in adults can result in a two- to five-fold enhancement of absorption for intrinsic iron or iron added in fortification of cereal products (Sayers et al., 1973; Bjorn- Rasmussen and Hallberg, 1974; Sayers et al., 1974a; Sayers et al., 1974b). It was suggested in a report of the International Anemia Consultative Group (1977) that a ratio by weight of at least 1:5 iron to ascorbic acid be used in infaht cereal and milk products that are fortified with iron. Some studies have indicated that ascorbic acid increased absorp- tion of food iron and iron in the ferric form, but had little advan- tage in iron therapy with ferrous iron (Bothwell et al., 1958). Reports by Brise and Hallberg contradicted earlier reports and found that ascorbic acid, when given in sufficient amounts (200 mg or more ascorbic acid with 30 mg of iron), increased the absorption of ferrous iron and that the absorption-promoting effect increased with increasing amounts of ascorbic acid. It was noted that individual variation was great between subjects and that such variation could have led to previous conclusions where no significant effect of added ascorbic acid was detected. It is thought that the reducing affect of ascorbic acid may help keep the complex in the ferrous state and thus prevent -~ or at least delay, the formation of insoluble or undissociated ferric compounds. Nojeim and Clydsdale (1981) have described the influence of pH in this phenomena. 28 It has been indicated that ascorbic acid promotes absorption of iron. Mechanism for increased absorption with vitamin C present may be an action via internal iron transfer systems. Mazur and coeworkers (1960; 1961) have shown a mutual relationship of ascorbic acid with ATP for the incorporation reaction of transferrin bound plasma iron into ferritin for 'Storage. Lockhead and Goldberg (1959) reported that ascorbic acid increases the transfer of iron to heme biOsynthesis as protoporphyrin. However, Brise and Hallberg (1962) concluded from their observations that the main effect of ascorbic acid under condi- tions examined (30 mg iron and 50-500 mg ascorbic acid administered orally) is intraluminal and probably due only to its reducing potential. Prospective for Combining_Ascorbic Acid with Fortification Iron in_ Infant Foods The concern regarding the persisting condition in the United States and in less developed countries with nutritional anemia has prompted attempts to fortify accessory food items with iron (Zoller et al., 1980). Development of iron-fortified salt, sugar, coffee, tea, oils and MSG have been accomplished. In view of the significant contribution for ascorbic acid in promoting iron absorption, attempts have been made in the past to ensure that both of these nutrients are generously supplied in a variety of foods. Sayers et al. (1974b) found that common salt could be fortified with both iron and ascorbic acid provided that starch was added to prevent development of dis- coloration. Sayers indicated that such fortification would signifi- cantly improve the iron nutrition in countries where the staple food is rice or maize. Such efforts would prove helpful in diets of adults 29 when iron nutrition is low and consumption of ironvrich foods or of traditionally~ironefortified foods is less than desirable. However, a proposal for the use of ironefortified salt or sugar in baby food is not consistent with current practice (Dallman, 1980) that includes recommendation for reduced salt and suagar in infants' diets. A more practical approach to improving infant nutrition is through enhancement of iron absorption from foods already providing significant calories and iron in babiesb diet. The ability of ascorbic acid to enhance iron absorption from ironefortified cereals, provided that it is added after cooking or taken during the meal as-a solution,as in orange juice,has been demonstrated (BjorneRasmmssen and Hallberg, 1974; Callender and warner, 1968; Elwood et al., 1978 ; Kuhn et al., 1968; Steinkamp et al.,1955). Cereal foods are promoted as standard dietary ingre— dients and thus may be an excellent vehicle for improving iron status in infants. The efficacy with ironefortification in the presence of ascorbic acid in infant cereal is clear; but the combination of the two nutrients in a single food item is feasible only if the appearance, taste and nutrient content is maintained until the time of serving . Additionally, the package distributed to infants as part of the national WIC (Women, Infants and Children Supplemental Food) program'includes both an ironefortified cereal and a fruit juice with ascorbic acid. The opportunity for the two nutrients to be served in a combined product is provided to a population who may be marginal for iron nutrition. Interest is therefore generated for investigating the chemical interaction of the combined nutrients. 30 Aacerbic 42151 .. Structure and Paorertise The Three Forms Idendified Reduced ascorbic acid (ReAA) (Figure l) is a highly soluble, acid compound with strongly reducing properties. It is a white crystalline solid melting at 192°C. It is the'Y —1actone of an hexonic acid with an enediol structure of carbon atoms 2 and 3 (Tannenbaum, 1976). The vitamin is readily labile to heat, low-acid pH, presence of oxygen, and metal catalysts. Dehydroascorbic acid (DHA) (Figure l) is formed with the removal of two equivalents of hydrogen from the reduced form. The oxidized form retains. the basic ring structure but no longer contains a conjugated system. Theacidity associated with the enediol structure of RrAA is lost on oxidation to DHA (Tannenbaum, 1976). Further oxidation results eventually in formation of diketo- gulenic acid (DKGA) (Figure 1) with the primary hydroxyl group being oxidized to a carbonyl group. Additionally, the completely oxidized form loses the lactone ring structure and subsequently all vitamin C activity (Tannenbaum, 1976). Influence 2£_Acid ponditions gaggeactivity RrAA possess two acidic hydrogens with pKa‘s in water of 4.25 and 11.79. The influence of pH on oxidation rate is marked and has been reported by numerous researchers (Khan and Martell, 1967; Lee et al., 1977; Joslyn and Miller, 1949). Below pH 6.0, the undiss— ociated and monovalent forms of ascorbic acid are the main species in solution (Khan and Martell, 1967). In this range, only the monOv ionic species was found to be reactive toward molecular oxygen and 31 ASCORBIC ACID O 3 2 c ‘ t -ofl r C = O / II 0/ I 0‘ \ C C " 0' C C = O n’ l H/ I Mil IO? U éItoII “I0“ L-ASCORBIC ACID L-DEHYDROASCORBIC ACID 0 II C C ‘-'-‘ 0 no. I C C =0 l I H c -OH OH I “lo-OH 2.3- 'DIKETOGULONIC ACID Figure 1. Three chemical forms of ascorbic acid produced during oxidation. 32 the spontaneous rate of oxidation varied linearly with the concentration of ascorbic acid. The contribution of the neutral species was essentia ally zero. Finholt et a1. (l963),and Blaug and Hajratwala (1972) found the rate of oxidation to reach a maximum near pKa1 of ascorbic acid. Joslyn and Miller (1949) determined that autoxidation of ascorbic acid in the presence of sufficient cyanide and thiocyanate to supress the catalytic effect of metals at pH 4.7 to 9.2 in phenol- sulfonate buffer (pH 7.8 to 9.2) and phosphate buffer (pH 4.7 to 7.6) proceeded with the divalent ion reacting 105 times faster than the monovalent ion with atmospheric oxygen. Under partial pressure, the reaction rate for the divalent ion is reduced onevfifth but is not reduced for the monovalent species. Influence with Presence 9; M§F§;§§F31y§F Joslyn and Miller (1949) reported that in the absence of thio— cyanate and in the presence of copper, the reaction rate for the monovalent ion was proportional to the oxygen concentration, yet the effect was observed only when the metal was present in high concentration. At low copper concentration, rate of oxidation increased more rapidly than did concentration of the metal. It is reported that a complex formation occurred between the ascorbate ion and the metal. Khan and Martell (1967) supported their results and found linear variation in rate with concentration of ferric (III) ions in the range of pH 1.5 to 3.85. In both cupric and ferric catalyzed reactions, the.rate showed an inverse dependence with the hydrogen ion concentration, but only at pH below pKa of ascorbic 1 acid (Figure 2). At pH values greater than pKal, there is an 33 14 12 .5 C3 ‘1 0V3 OD KMINX1 Cu (II) x 105 Figure 2. Catalytic effect for the oxidation of ascorbic acid in the presence of Cu(II) ion at 25 C, as function of pH. K - difference between the first-order rate constants in the presence and in the absence of the metal ion. 34 increasing proportion of divalent ions in solution. Joslyn and Miller (1949) reported that the rate of oxidation increased to maxima between pH 6 and 7. This finding may reflect some spontaneous oxidation with the increased rate observed for the divalent ion in reaction with atmospheric oxygen. The relationship with pH and metal catalyst is complex and is influenced by a variety of factors including buffers in the system and amount of oxidation already taken piace. During the course of oxidation, pH value changes somewhat as the less acidic DHA accumu- lates. Joslyn and Miller (1949) observed an increase in pH given pH values of 6.5 and below; but at pH above 6.5, a decline in final pH occurred, indicating that products of oxidation other than DHA must accumulate. Despite varying rates of ascorbic acid degradation reported over time, researchers including Lee et a1. (1977) have shown a firsteorder reaction rate for ascorbic acid concentration versus .storage time. This has been reported for anaerobic conditions (Lee et al., 1977; Finholt et al., 1963) and for aerobic conditions (Khan and Martell, 1967). Influencejog;StabilitygbthaterjActivity Mbch interest has been generated by investigators for the relationship between loss of food nutrients and moisture content. The moisture content is most appropriately represented as water activity (aw) defined as; (1) 35 _ where Po F vapor pressure of pure water, P a vapor pressure of water in the food product and Z ERR = percent relative humidity of the system in equilibrium, whereby there is neither gain nor loss of water. The majority‘of interest with water activity is focused in dehydrated foods or intermediatesmoisture.(IMF) technology. Nutrient stability of a system With low water activity is quite different from one in a‘more-aqueous state. Karel and Nickerson (1964), and Vojnovich and Pfeifer (1970) have reported that in foods of water activity up to aw~' 0.5, there is an increased rate of ascorbic acid destruction with. moreasing‘moitsture content. Lee and Labuza (1975) reported the influence of aw on destruction of vitamin C in systems with aw up to 0.84 and found that the.same phenomenon existed whereby aw of a food system corresponded with increased rate of ascorbic acid degradation (Figure 3). Threeamechanisms were proposed by which water may act to control reaction rate. First, the mechanism for oxidation may differ at various awFs; second, that water may act to dilute the concentration of ascorbic acid which could reduce the rate of reaction. Alternaa tively, an increased water content may facilitate the reaction since higher.moisture corresponds with_a less viscous solution and enables 'more.diffusion to take.place. Analysis by Lee and Labuza (1975) concluded that oxidation rate does not change assw increases and a dilution effect may be occurring. However, in their studies, it appears that the-dilution effect may be masked by some other mechans ism that caused the reaction rate to increase. The increased mobility of a reactive species in the more aqueous solution was considered as a most likely agent for enhancing oxidation rate. The increased FRACTION ASCORBIC ACID REMAINING 36 24 48 72 96 TIME (hrs) Figure 3. Retention of ascorbic acid as function of aw at 35 C. 120 mi f i th be by 32 rate could be due to either increased mobility for the metal catalyst or for ascorbate itself. It was further postulated, however, that above a certain moisture content, the aqueous phase viscosity was not greatly altered and the rate of oxidation should approach a constant value. Methods 2:.ABQLY3§§.fEE AsCorbic Acid Numerous analytical procedures exist for ascorbic acid in foods, yet none are completely satisfactory since foodstuffs contain inter— fering agents and some processing techniques characteristic for a food item contribute interfering agents (Freed, 1966). Analysis is generally by means of a procedure based on an oxidation—reduction reaction, or by means of treatment for a chromagen formed by coupling oxidized ascorbic acid with 2, 4vdinitrophenylhydrazine dye (DNPH). Classic analysis follows oxidation by a redox dye, such as 2, 6- dichloroindophenol (DCIP). T12 DNPH techniqueuaybe used to determine only the reduced form of ascorbic acid, or may yield "total" vitamin C which.includes both active forms: reduced form, R—AA and DHA. DHA is reduced to R—AA, and DKGA is quantified by the amount of dye required to form a highly colored hydrazone. The amount of total ascorbic acid present is indicated by difference. The three forms of ascorbic acid may be quantitatively deter- mined using a stepdwise reaction with the compound DNPH in a method first outlined by by Roe et a1. (1948). The analysis is based on the coupled DHA and DKGA derivative. The reaction is known to be quite sensitive (Penney and Zilva, 1945). The technique discussed by Roe et a1. (1948) is timeeconsuming and laborious, however, 38 methods;have.been.out1ined. for various particular foodstuffs. The coupling reaction is delicate and temperature is a critical factor. At temperatures in excess of 37°C, the reaction is affected by glucose, fructose, and glucuronic acid (Roe, 1961). At low temperatures, the reaction is quite slow. Extraction of the three compounds from tissues must be conducted so as to prevent the oxidation of RsAA to DHA, and to keep at a minimum the change of DHA to DKGA. Roe and Oesterling (1944) prevented such oxidation in plant tissue by using a metaphosphoric acid solution containing 1 percent thiourea. In animal tissue, a more powerful antioxidant agent is required due to the strength of the oxidizing affect of oxyhemoglobin (Roe et al., 1948). The problem was solved by adding stannous chloride in 10 percent concen« tration to the metaphosphoric solution, then diluting the sample to contain 0.5 percent SnClz. Cereal and fruit foodstuffs may not require presence of SnCl2 as antioxidant, but iron added in fortification may become an inter? fering agent and catalyze the oxidation of ascorbic acid. Thus, the.use of SnCl2 may be recommended for analysis of ironvfortified products. Furthermore, Mattews and Hall (1978) compared ascorbic acid in fresh and frozen green peppers using thiourea and SnClz. Researchers determined that RsAA.was protected from oxidation to a greater degree with SnCl than when thiourea was used. It is there- 2 fore apparent that recommendation should follow for use of SnClz as antioxidant in ascorbic acid analysis of cereal and fruit infant foods when an antioxidant is required in the procedure. In light of difficulties inherent with the method proposed by 39 Roe et al. (1948) which includes use of several toxic reagents, alterna- tive means of quantifying the three forms of vitamin C have been investigated. The classic method for analysis of reduced ascorbic acid includes oxidation by DCIP. Oxidation of the two loosely- bonded hydrogen atoms of the dienolic group of ascorbic acid is generally measured by visual titrametric techniques using the indophenol reagent, yet these techniques are limited by interfering agents and by difficulty in determining the precise endpoint of titration. Photo— metric analyses have aided precision, but no satisfactory adaptation of the technique has been developed that avoids interference of compounds such as thiosulfate, sulfite, reductones; and ferrous, cuprous, or stannous salts which react rapidly with indophenol. DHA and total ascorbic acid (TAA) may be determined by a fluoro- metric procedure involving the reaction of DHA with orthophenylene— diamine (OPDA) to yield a fluorescent quinoxaline, 3- (l, 2—dihydroxy— ethyl) furo[3, 4vh] quinoxaline-l-vone (Deutsch and Weeks, 1965). Specificity of the procedure is enhanced through use of boric acid. Boric acid is added to the oxidized aliquot prior to the addition of OPDA, causing formation of a boric acidvDHA complex. DHA may not condense with OPDA and any resulting fluorescence is then due to extraneous material. Background interference may be subtracted from the fluorescence of the complex formed without addition of boric acid and true.DHA is indicated. Use of Norit as oxidizing agent for oxidation of all three forms of vitamin C present allows for the determination of TAA,and RsAA is attained by difference. A simple, rapid, quantitative method for continuous flOW‘ analysis for DHA and for TAA has been developed by Kirk and‘Ting (1975). The method, using the Technicon Autoanalyzer II, relies on 40 a modification of the manual method of Deutsch.and Weeks (1965). Oxidation of R<~AA is by DCIP-ethimrea oxidation in substitution for Norit; since NOrit oxidation is incompatable.with the continuous flow system. Use of the automated system decreases analytical time without adverselyraffecting either the accuracy nor precision of the OPDAvbased assay. EXPERIMENTAL PROCEDURES Preparation of Model Systems Two types of model systems were prepared. The intent was to simulate experimental products of (a) a mixed cereal system of dry cereal and apple juice and (b) a prepared wet cereal product. The mixed product, referred to as mixed cereal, was simulated with a 100 Brix sucrose solution, and the wet cereal was emulated with a 5 percent gelatinized rice starch solution. Rice flour was heated to gelation temperature and cooled thoroughly before addition of ascorbic acid. Both products were freshly prepared for use each day and were stored at room temperature. Ascorbic Acid petermination ReAA and DHA were determined by the continuous flow'ortho~phenylene— diamine micrvaluorometric procedure described by Kirk and Ting (1975). Ascorbic acid was extracted with 0.5 percent oxalic acid and samples were filtered before introduction to the Autoanalyzer (Technicon Autoanalyzer II, Tarrytown, N.Y.). DEA condenses with the OPDA dye forming a fluorcphor which may be detected fluorometrically with an excitation wavelength of 360 nm and emission wavelength of 436 nm. DHA'blanks are determined with addition of boric acid to prevent condensation of OPDprith DHA. Total ascorbic acid is measured as DHA following oxidation of R-AA with 2,69dichloroindophenol (DCIP) dye, ReAA may be determined by difference of DEA.from TAA once “respective blanks have been subtracted. A set of ReAA standards (Eastman KOdak) were run daily and ReAA was determined by applica— tion of a linear regression equation. DHA standards (ICN Pharmaceuticals) 41 42 were run simultaneously when DEA was to be determined and quantity of DHA was calculated similarly by application of linear regression. Reduced Ascorbic Acid Analysis for RrAA followed the classical titrametric technique described in AOAC (1975; Freed, 1966). Samples were prepared by extraction in 3 percent metaphosphoric acid (mHPO3) and 0.005 M.EDTA solution. Extracts were titrated with DCIP and concentration was determined visually by the amount of dye required to oxidize the ascorbic acid and produce a faint pink coloration. Dehydroasconbic Acid DHA was determined by the automated micro-fluorometric procedure described by Kirk and Ting (1975) and also by adaptation of the colorimetric method outlined by Roe and associates (1948; Freed, 1966). The method fOr automated analysis has been described above for RvAA, but DHA concentration was determined by application of linear regression formulas for DHA standards. The procedure outlined by Roe et al. (1948) permits identification of RwAA, DHA, and DKGA; but involves the use of hazardous reagents and is difficult to manage, Both hydrogen sulfide and bromine are required in the assay, which are corrosive and toxic; subsequently, a modification of the method outlined by Roe et al. (1948) was adopted. The analysis used is based on the coupling reaction of both DHA and DKGA with 2,4—dinitro- phenylhydrazine (DNPH) under carefully controlled conditions to give red- colored osazones, DHA content in model systems and in samples is deter- mined after comparison of color produced in samples and a set of DHA standards. Analysis for baby food samples would indicate both DHA and DKGA 43 concentration present. Preparation for each.sample is by grinding in a mortar with.l/20th.of the final volume af 5 percenthPO3 and sufficient dry SnCl2 to make a 10 percent SnCl solution in the volume of 5 percent 2 mHPO3 added. A 10 percent concentration for the SnCl2 grinding 'protects ascorbic acid from oxidation and leaves DHA and DKGA solution during unaltered. A 20—fold dilution of samples follows grinding to yield the final extract. Specifically, to each sample 0.5 g SnCl2 was added with 5 ml mHPOB, samples ground and diluted to 100 ml. Extracts were subsequently filtered; 4 ml aliquots pipeted into each of three test tubes. One set- of tubes served as blank, while other tubes were treated with 1.0 m1 of 2 percent DNPH. Tubes were placed in a 37°C water bath for exactly six hours to allow the coupling reaction to proceed. Samples were immersed in an ice bath to cool, and 5 ml of 85 percent sulfuric acid was added drop- wise to each tube with continuous shaking. To the blank tubes, 1.0 m1 of 2 percent DNPH was added and all tubes were permitted to stand at room temperature for exactly 30 minutes. Since color intensity is critically dependent on time following addition of H 804, a definite schedule was 2 followed for spectrophotometric.measurement of color. Tubes were allowed 7 to stand 30.minutes and read in a Bausch and Lamb Spectrophotometer (Spectronic 70, Bausch.and Lomb, Rochester, N.Y.) at 520 nm. Evaluation at 520 and 540 nm revealed that maximum absorbance occurred at 520 nm. Concentration of DHA was evaluated by comparison with a set of DHA standards and application of a linear regression model. DiketoguloniC*Acid DKGA in samples:was determined by the method described by Roe et al. (1948). The colored osazone is the .COuplsicomplex of DKGA with DHA. The completely oxidized component DKGA was determined by difference, 44 subtracting DHA determined by the automated analysis. By using the autOv mated analysis for determination of DHA, the use of corrosive and toxic reagents was avoided. mos Eager mess-3 Comparison of Methodolcgy for R335 Recovery of RrAA from both sucrose and starch model systems was evaluated simultaneously by the autbmated analysis and by the titrametric procedure. Samples were prepared with RrAA standard at concentration of 100 ug/ml. Samples prepared for automated analysis were further diluted ‘with 0.5 percent oxalic acid to contain 4, 6, and 8 ug/ml R-AA. Starch solutions were filtered before introduction to the Autoanalyzer; sucrose solutions did not require filtration. Blanks for model systems were also evaluated by preparing samples with no ascorbic acid added. Samples were prepared for analysis fer the titrametric procedure by diluting with 3 percent mHPO e0.005M EDTA solution to contain 200, 400, and 600 ug/ml 3 RrAA. Differences in sample preparation between the two methods corres- ponded with sensitivity for each of the two methods of assay. Blanks of both model systems were prepared as for the automated analysis to evaluate background interference. Comparison of Methodology gggwgflé .Recovery of DHA from both.sucrose and starch model systems was evaluated by the automated analysis and by the spectrophotometric proced- ure, Samples for both systems were prepared to contain 100 ug/ml DHA. Samples for analysis using the Autoanalyzer were diluted with 0.5 percent oxalic acid to contain 4, 6, and 8 ug/ml DHA. Only starch samples were filtered before introduction to the systemg. Blanks for both sucrose and 45 starch solutions were run simultaneously with samples, as were DHA.standards. Samples prepared for evaluation by the spectrophotometric assay with DNPH required addition of 10 percent SnCl2 as antioxidant, grinding with mortar and pestle, and dilution with 5 percent mHPO to contain 4, 6, and 8 ug/ml 3 DEA. Concentration was determined by comparison with a standard curve. Standards were prepared in either oxalic acid or mHPO respectively. Blanks 3 were run for both sucrose and starch solution in both automated and spectro~ photometric analyses. Exsparatiss 2i infest .Food 39913193 Two products were evaluated. These were a mixed cereal: dry rice cereal with bananas prepared in a 1:5 dilution with apple juice, and a wet cereal: rice cereal with applesauce and bananas. The wet cereal is a thermally processed product packed in glass jars with a net weight of 100g. All samples including the dry rice cereal, apple juice,and wet cereal were prepared by Gerber Products (Fremont, Michigan). Rice cereal is commercially prepared with electrolytic iron added in fortification to make label declaration of 47.5 mg/lOOg. Considering a serving size of 14.2 g (0.5 oz), cereal may provide 45 percent the U.S.R.D.A.(c°de‘of Federal Regulations, 1981) infants. Apple juice is prepared to enable label declaration of 32.0 mg reduced ascorbic acid per 100g (3.2 fluid oz), a level that represents 120 percent the U.S.R.D.A. for infants and 100 percent the U.S.R.D.A. for a child one to four years of age. Complete nutritional information for the three products as prepared by Gerber is given in Table 1. Rice cereal was also prepared with no iron added in fortification. Unfortified cereal was prepared with apple juice identically as was product containing electrolytic iron. Three batches of each mixed product with and without iron were prepared in sufficient Quantity to enable 46 Table 1. Average nutrient values for baby food products per 100 g. Product Rice cereal A la juice withe cel:::uce nutrient with bananas PP and :Zganas Calories 396 49 81 Protein (8) 9.1 0.1 1.3 CEO (3) 75.6 11.9 18.1 fat (3) 6.4 0.1 0.4 crude (8) 1.0 0,1 0.2 fiber moisture (8) 5.1 87.6 79.7 Ca1cium.(mg) 634 4 10 Phosphorus (mg) 352 5 23 Iron (mg) 47.5 0.4 5.0 Sodium.(mg) 100 2 5 Potassium (mg) 660 93 49 Vit A (IU) 29 - l thiamin (mg) 1.58 - 0.17 riboflavin (mg) 1.90 0.01 0.20 niacin (mg) 14.08 0.08 2.67 v1: 86 (mg) 0.63 0.03 0.13 Vit C (mg) 2.1 32.0 11.7 47 sampling over a 10 day period. Samples were stored at 4.4OC in covered plastic containers and were held in the dark between sampling. Two batches of vitamin C—enriched juice and two batches of ironefortified and unfortified cereal were provided by Gerber. Three different combinations of the mixed product, both with and without iron, were prepared as diagramed in Figure 4. wet cereal, rice cereal with applesauce and bananas, was prepared as for commercial distribution fortified to a level adequate to make label declaration per 100g of 5.0 mg iron as ferrous sulfate, and enriched to declare 11.7 mg reduced ascorbic acid. Based on serving size of 135 g (contained in a single serving jar), this corresponds with 45 percent of the the U.S.R.D.A. for infants for both iron and ascorbic acid. Products were prepared by Gerber both with and without iron—fortification, but with vitamin C—enrichment included in all products evaluated in the experi- ment. Three batches of each product, both with and without iron added, were prepared in sufficient quantities to permit sampling over the experimental period. Sampling from each replicate was conducted from a single jar. All jars were stored with refrigeration between sampling at 4.400 and in the dark. Schemmatic diagram of all samples as prepared for analysis is given in Figure 5. Both mixed cereal and wet cereal samples were prepared for introduction to the Autoanalyzer for determination of RrAA and DHA, and for spectro— photometric analysis for DHA and DKGA. Dilution to the appropriate range of assay required that 1.0 g of mixed cereal or 3.0 g of wet cereal be diluted to 100 m1 with.0.5 percent oxalic acid. Extracts were filtered before introduction to the systemi Identical samples were prepared for spectrophotometric analysis. Range of the spectrophotometric assay necessitated that 2.0 g of mixed cereal be diluted to 200 ml with 5 .coauuumooua Hoouoo pox“... no 55393 usumaosom .e ousmah :03): :0: o: o o 5200 >5 , .oocoo >5 :03}: cos 0: coax; . 0: Q n— w m co: 30.00 in. $0.00 to $0.00 in 30.00 in 4 8:2... m: 5.5.... m: . o 5523} . o 5525;, N 02.2. _. 00.2. mum-m. 30:3. 059:3 8.. 5.; $0.00 03 .mumhaooo you wouoooue monnaom .m change E- <- E/ :2. 03203020 $0.00 0055. mm :o.__ o: :9... 5.3 :0: o: 50 percent mHPO and that 2.0 g samples of wet cereal be diluted to 100 m1. 3 Iron Analysis Iron assay was conducted for all samples at the beginning and end of the 10 day experimental period. Analysis was by atomic absorption spectro- photometry. Sample preparation was conducted using a wet ashing technique and digestion with perchloric and nitric acids. Samples for day 0 of the experiment were stored until the conclusion of the time period with the addition of 10 ml nitric acid before storing. Digestion was completed upon termination of the experiment and all samples were evaluated simult- aneously. An atomic absorption spectrophotometer (Model IL 951 - Instrumentation Laboratory, Inc., Wilmington, Ma.) was used. The model uses a premixed gas burner. Fuel used was acetylene with air as oxidant. Light source was a. hollow cathode lamp (8 ma). Absorption of radiation was detected at a wavelength characteristic for iron, 248.3 nm, with a band width of 0.3 nm. Sample concentration was determined by integration with a set of iron standards within the predicted range of samples being evaluated. Data Assists Results of the pilot study comparing methods relative to standards were compared for statistical differences by means of a threeeway analysis of variance (ANOVA). Analysis was run with the SPSS computer program (Statistical Package for the Social Sciences). The less of ReAA and TAA, and the accumulation of DHA and DKGA was evaluated in terms of actual condentration change by means of the SPSS computer program. Differences were first described based on three-way ANOVA. Evaluation followed according to the repeat measure spliteplot 51 design for the experiment (Appendix IéIII,‘V). A first—order reaction rate was assumed for loss and accumulation of the respective forms of ascorbic acid: ‘vaaécz _ = k.C (2) where (C) a concentration of ascorbic acid; t = any time during storage (days); and k - firstrorder reaction rate constant (daysQl). The k constant was calculated for each system, both mixed and.wet cereal products, and for each form of ascorbic acid: TAA, RvAA, DHA and DKGA in terms of concentration ratio: concentration ratio = c / c0 (3) In cases where negative values would have resulted in calulation for equation 3, another form of this expression was applied: concentration ratio = c a co / co (4) The k constant is useful in that it can be applied to calculate concen— tration (c) at any time (t) given knowledge of initial concentration (co) by means of the firsteorder reaction equation: c = co exp (ear) (5) The k constant was also used in calculation of oneehalf life, or doubling time,for each form of ascorbic acid. This would indicate the time required to reduce by oneehalf the amount of TAA or R—AA present, 52 or to double the quantity of DHA or DKGA in the mixed and wet cereal products. seedssstsl ...ssDesi Based on the considerable interest expressed regarding the persisting condition in the U.S. and in less developed countries with nutritional anemia in infants.(Fomon, 1970; Filer, 1969), and the increasing interest in iron--fortification with highly available forms of iron, along with the potential for interaction of nutritional components enhancing bio- availability (Sayers et al., 1974), a project was designed that would assess the effects of the two forms of iron commonly added in infant foods. The foods selected have been serving as vehicles for both iron— fortification and vitamin Cvenrichment. That ascorbic acid enhances iron bioavailability has been established (Apte and Venkatachalam, 1965; Layrisse et al., 1968); yet, if iron in combination with ascorbic acid acts to degrade the vitamin then its addition may not be acceptable in infant foods that contribute substantially to the vitamin C nutrition of infants. Electrolytic, or elemental iron is currently added in fortification of dry infant cereal which may be prepared with apple juice for infant feeding. Reduced iron, ferrous sulfate, is added to wet cereals that are simultaneously enriched with vitamin C so that they may provide substantial quantities of both nutrients in infants' diets. Reduced iron may possess greater catalytic capacity to oxidize ReAA since it is chemically a more reactive species of iron than that in the elemental form. 53 Pilot Projects for, Comparison of Methodology A study was conducted preliminary to the project designed to evaluate accuracy in each model system relative to standards for the methods used in the experiment. The combination of methods with the automated analysis for determination of TAA, RvAA and DHA, and the spectrophotometric lanalysis for quantification of the coupled DHA.and DKGA complex was intended to permit evaluation of DKGA by difference from the DHA component. Since methods were not quantitative, the DKGA component could not be identified. Correction factors were required that would permit comparison of results obtained to standards in model systems. Reduced ascorbic acid standards were run for the automated analysis and the classical titrametric technique. DHA standards were run for both the automated analysis and for the adapted method from Roe et al. (1948). Sample size for the two comparative studies was predetermined to obtain statistical power greater than 0.95 for an f-test; analysis of variance of fixed effects. traps-Day .83: press as: £93: :31:- Esreal ‘ng ans {cede The project was then initiated in order to enable comparison for catalytic reactivity in destruction of vitamin C for mixed cereal with electrolytic iron, and wet cereal with ferrous sulfate. Since sampling was to be conducted daily from each sample over a 10 day period, a non— random, splitsplot, repeat measure, experimental design was applied. Each sample may be treated as its own block, thereby reducing need for excessive replication of product. The design obtained adequate statistical power with three replicates of each product, both with and without iron. Experimental error (mean square.error) was reduced without sacrificing power of significance. The splitvplot, repeat measure design is summarized 54 in Figure 6. The design is useful for indicating trends over time. as response to a given treatment (Gill, 1978). Bonferroniet analysis and a Student's t—test were also applied to make pairvwise comparisons between combinations of cereal products. Indication gf_Background Interference‘wit§\A§corbic Acid Samples were prepared by Gerber identically with experimental product but with no vitamin C added in enrichment. Analysis was conducted similarly forthe unenriched product as for the enriched samples to indicate the amount of natural vitamin C in each product. Iron Evaluation Each product was evaluated for iron content at the beginning and end of the 10 day experiment. This was conducted to verify that iron was present at levels declared by Gerber and that iron content was not changing dramatically during the course of the evaluation. Data evaluation was conducted according to the Student’s t distribu— tion for pairewise comparison. Change in samples without iron added was compared over time, as was change in samples with.iron added in fortification. Results are expressed in terms of significance with a t-test statistic (Appendix VII). 57 SPLIT - PLOT DESIGN (repeat measure) trt (3) d.f. CT 1 IRON 1 CT X IRON l batcheleT X IRON .8 DAYS 9 CT X DAYS 9 IRON X DAYS 9 batches X DAYS .12 Figure 6. Split-plot design of the experiment. RESULTS AND DISCUSSION The study of nutrient interaction within a food product is particularly significant when considering the infant's diet, since the consequence of suboptimal nutritional status during development may influence health in adulthood. The effect of two forms of iron added to infant food on ascorbic acid was evaluated. The two products evaluated represent vehicles with substantial nutrient contribution" in infants' diet, and are currently significant sources of iron and ascorbic acid for infants fed commercial cereal foods. The systems were chosen to represent a realslife feeding situation and thus, were not manipulated to equate iron level nor ascorbic acid content. Nutritional implication for feeding each product could be assessed. Comparison‘gijethodologyjfgrugzéé Results obtained from the automated method were compared with that of the classical titrametric vitamin C ana1YSis. Identical readings were not obtained for each method. Comparison for the 106 Brix model solution in relation to ReAA standards is illustrated in Figure 7. Both methods indicated less ascorbic acid than was present by addition of amounts of R-AA standard. Results for the titrametric method were significantly below that obtained for the automated analysis. Comparison for methods with the starch model system indicated a similar but reversed situation. Both methods indicated less RvAA than was present by addition of the standard; but in starch, the titrametric assay indicated a greater amount present thanwas indicated with the automated analysis. This difference was. significant (p( .001) and is 57a 58 .mvosuoa owns—05.5%» vow Houhdmomousu hp ovum—.6000 1.... fr. .8332... H208 5.0.. 62 fl .2... 08 8389.80 .e 88...... D¢40+2m N12 7 OPDA ‘ .9. DHA ’ a. - 6 CH (0H) CHZO-I 5 N H 3-(l.2-DIHYDmXYE1‘HYL) Fl]RD[3.4—B] mImMLIME-lnt Figure 9. Formation of a fluorescent compound. with DHA that resulting from interfering compounds. Boric acid, used in the analysis to account for interference from compounds other than DHA, complexes with adjacent gi§_hydroxyl groups of carbohydrate (Malcolm et al., 1964). The diol groups of the furanose structure of DHA bind with boric acid to form monoé, neutral and bisdiol complexes (Figure 10) (Pigman and Wolfram, 1949). Once complexed prior to addition of OPDA, DHA does not condense with indicator dye. Therefore, any resulting fluorescence should result from extraneous material. However, Dentsch and weeks (1965) indicated that a compound posessing all the following qualities could interfere with theaccunacy of the fluorometric method: (a) an‘d,ediketo group which reacts with the-diamine dye (b) fluorescence wavelength of the formed quinoxaline within the 62 region of the.assay (c) contain cis hydroxyl groups which react with boric acid solution to form a complex. ‘ | 'C-—0\_/0H -C"—""0\ _ B I / B ‘OH -C--0/ \Qi -C—--0 I l WIOL BORIC ACID NEUTRAL -"c———o\_/o—— ,- BISDIOL BORIC ACID Figure 10. Borate—carbohydrate complexes. It is not likely that the observed result stems from a binding of boric acid since this would result in detection of fewer extraneous substances and would increase rather than decrease the observed amount of RvAA. Smuld any agent be active in inhibiting the borate-complex _formation, then a larger factor would be subtracted from the total fluorophor perceived as oxidized DHA. This could account for the low results obtained in this comparison study. An additional contributing factor to the low observed result could be incomplete oxidation of reduced ascorbic acid to DHA.by DCIP. Therefore, the amount of available DHA bound as the fluorescent complex would not represent all ReAA present. Although result of the comparative study in starch was reversed relative to the result obtained for the sugar model solution, both methods 63 indicated a lower concentration of RaAA than was added to each solution as a standard. This observation reveals a similar phenomenon as that which occurs in sugar, that is, there is interference of some nature. Natural starch is comprised of glycosidic sidechains and yields only glucose on hydrolysis. Glucose is a reducing sugar and upon hydrolysis, would interfere with accurate determination of RsAA present. It may follow that the fluorometric method is less affected in starch than sucrose, or that the influence of the model system affects the titrae metric method disproportionately. Comparison of Methodology EQEHQEE. Results obtained from the automated analysis were compared with those of the spectrophotometric assay for a set of DHA standards. Again, readings obtained by the two methods were not identical, and neither method yielded results as would be expected by known amounts of DHA standard which was added to the sugar and starch model solutions. Results for the spectrophotometric analysis for the 100 Brix model system recorded amounts that were significantly higher than known amounts of standard added (p‘<.001), and were significantly higher than were indicated by the automated method C p‘1 .001) (Figure 11). Roe (1961a) reported that the coupling reaction of DNPH method is not sensitive.to reducing compounds which interfere with the oxidation- reduction techniques, Only at temperatures that exceed those advised for the coupling reaction would there be significant interference from sugars. Sugars resulting from possible hydrolysis in the sucrose model solution may have appeared as interfering agents for the titrametric analysis for RmAA that was reportedly sensitive to such compounds. Without the interference present in the DNPH assay for DHA, observed .moosuoa aquuoaouonnouuoonn o5 nonhuoooouom h.— noun—Vanna 33 fit. 8338 ~89... 5.8 03 5 <5 you 83838 .2 253m oca4<0 v N K. .. / 9-0) OLLVH NOLLVULNBONOO I O ~o.”° (°o 80 period was significantly greater than for the same product without irons fortification. DHA for wet cereal with iron was only slightly greater than for the wet cereal product without iron. This distinction may be the result of a definently enhanced oxidative ability of electrolytic iron compared to that of ferrous sulfate, or may be the result of conditions imposed on both nutrients, iron and ascorbic acid, within the two cereal products. Figures 14 and 15 indicate that the difference in DHA.when iron was present was more.dramatic in wet cereal (Figure 15) than in mixed cereal (Figure 14). The observed result appears to support the concept that cereal type had a pronounced influence on the catalytic effect of iron added to the product in fortification. Figures 19 and 20 more clearly show*the distinction between non- fortified and fortified products. Concentration for DHA in fortified product is presented as a ratio of DHA in non—fortified cereal. Figure 19 illustrates for mixed cereal that an increase in DI-IAwasevident after day 4 of storage. The increased ratio for fortified product continued until the conclusion of the 10 day period, No trend was observed that could indicate a gradual increase in DHA, and the difference between non—fortified and fortified product was not altered after the single substantial increase on day 4. Figure 20 presents the parallel situation for wet cereal product. No trend was observed and results do not reveal any significant differences. There was a greater concentration.fin:DHA achieved in the ironafortified product on day 6, and a ratio exceeding. unity in favor of the fortified product continued until the end of the 10 day period. The difference between the wet cereal products was not significant, whereas the mixed cereal did show a significant effect for 81 0.. .Auooooum ooamfiuuomlsos "oo«MHuHom Isoua mo oauou mo oommouaxowfloouoo ooxfia ooamauquIoouw our ooauwauomusos you oquou soauouucoosoo an: o« owsonu .mA madman m><0 0 m V (4.. . [ 30. V 0. 22.. 1.23 .H. 22.. oz I <10 ..r OILVH‘ 82 0r .Auooooue ooumauuomtsoo "roameuuom loops mo oauou no oommoumxov Hoouoo nos ooamauuomlsoua ooo ooAMHuuowlso: you oauou doauouusooooo <0 0 .h 0 m w 0 N w mz 20:. at? U 22... oz I <10 06 Q& 83 DHA.when iron was present. Diketogulonic Acid_ Results for the completely~oxidized component, DKGA, were more erratic than for the other forms of aScorbic acid. All factors influenced DKGA, but it appears that they do so independently. There was an overall increase in DKGA during the 10 day storage period (pl: .001). Also, the average concentration of DKGA was significantly greater in wet cereal than in mixed cereal (p< .05). These results support the notion that the cereal product itself may predominate in influencing ascorbic acid. The influence of iron was significant for DKGA (p‘< .001), but the iron factor failed to interact with either cereal type, i.e. the influence of iron was similar for both cereal types. Statistical eval— uation for DKGA is presented in Appendix V and Appendix VI. ‘VT‘V Firsteorder reaction rate values, given in Table 3, are graphically illustrated in Figure 21 for mixed cereal and Figure 22 for wet cereal. Negative reaction rates-for TAA and ReAA reflect loss of these components omurtime. Alternatively, positive values for DHA and DKGA reflect the accumulation for these components over the-course of the experiment. Although the concentration ratios determined for TAA and RfAA begin at 1.0, results have been normalized so that values could be described on the same axis with DHA and DKGA, wflflxall’values originating at zero. The significantly greater rate of DHA accumulation in ironvfortified cereal relative to the nonefortified product is apparent in Figure 21 (p'< .05). IronefortifiCation with electrolytic iron does not cause 84 Figure 21. k First-order reaction rate for ascorbic acid metabolites in mixed cereal. DKGA I: 0 .- 0 I: sacs < I .t 1 I: I ‘3 a 0 g |. < o z a C o a I r. 3 POP ouvu Mouwiuaonoo 98 “2qu Time 1EMA-- O 86 Figure 22. k First-order reaction rate for ascorbic acid metabolites in wet cereal. 3.0 °. N DKGA DHA O 0 N ouvu nouvumaouoo [.8 °. ' Dl-iA Time R-AA 1.0 88 a greater rate of loss for TAA or RqAA, nor a greater rate of increase for DKGA compared to the nonefortified mixed cereal. Figure 22 illustrates the similar situation for wet cereal; however, there is no significant difference between reaction rates obtained for ascorbic acid metabolites with the addition of ferrous sulfate in the wet cereal product. When k values are contrasted for cereal products, there is a significantly greater rate of DKGA accumulation in fortified wet cereal than in iron? fortified mixed cereal (Appendix VI). No other ascorbic acid metabolite shows significantly different reaction rate between the two non—fortified cereal types or the two ironefortified cereal products. Differences in k are further illustrated in Figure 23 for loss of TAA, Figure 24 for loss of RuAA and Figure 25 for DHA.increase. The con— centration ratio for each cereal type has been averaged for both non- fortified and iron—fortified product, and is presented to indicate change in concentration ratio over the 10 day storage period. The DKGA component failed to show significant differences in the averaged concen- tration ratio over time for the two cereal products, and is thus not indicated. The most dramatic difference is for DHA, with wet cereal having a markedly greater rate.of increase in concentration ratio over time than for’mixed cereal (figure 25). An additional means of comparison for each metabolite of ascorbic acid evaluated in cereal products is expression of one—half life. In terms of RvAA and TAAg onerhalf life represents time (days) required to reduce by onevhalf the concentration present on day l; and for DHA and DKGA, a parallelism.of the expression amounts to the time necessary to achieve a tweefold increase from the initial concentration. Half-life, or doubling time (in days) for all products is presented in Table 4. Ratio for change in onerhalf concentration, or tw0¢fold concentration, is 89 .Amuoavoua vowwauuomlcoua can vowmwuuomlao: Mom wwwaum>mv ouoavoum awoken um: was evade n«_<uv 3369.3 Hmouoo no: can v33... a.“ film you muaausou mummie— . «N 0%.me Op 0 m>uv nauseous Hmuuuo pus can mouse as OPDA > titrametric Starch: standard) titrametric) OPDA RFAA Sucrose: spectrophotometric >> OPDA» standard Starch: OPDA? spectrophotometric > standard For ReAA in the.sucrose model solution, the automated OPDA procedure yielded results that more closely agreed with known amounts of standard added than did the titrametric procedure. Results in starch were inverted and the titrametric assay produced more precise recordings relative to a 109 110 set of standards. For DHA, the pattern of results was similar with the automated OPDA analysis indicating closer agreement with standards relative to the spectrophotometric method in sucrose; but in starch the spectrophoto- etric method was more accurate. Results for DKGA, obtained by difference, were not quantitative between methods. A correction factor, relative to a set of standards, was required to allow comparison between products in the experiment. The three inconsistencies in methodology described leave the researcher in question regarding the precision with some techniques used for ascorbic acid assay. There appears to be considerable differ- ences for ascorbic acid concentration determined by some methods of analysis. Magnitude and direction of the inaccuracy depends largely on characteristics of the food system evaluated. Further comparative research may validate the observations reported herein that reveal inconsistency between methods currently used to determine ascorbic acid concentration, and are commonly reported in literature. It was postulated that iron in the reduced ferrous form, present in wet cereal, would enhance the rate of oxidation relative to that for elemental iron in mixed cereal. This assumption was not supported by experimental data. Significant differences were obtained for DHA between nonvfortified and ironefortified mixed cereal during the course of the storageperiod; however, there was not a corresponding significant distinction between wet cereal products. The pattern was apparent in both cereal products with a delay in the initiation of the oxidative process in ironvfortified product until day 4 or day 6 of the experiment. The longer delay was observed for the wet cereal product. This similarity between cereal types is supported by 111 work of Nojeim and Clydsdale (1981) who observed a conversion of elemental iron to ferrous form in 48 hours. With.such conversion, there is minimal nutritional significance given the product is stored for a period longer than 2 days, since iron in both products would be present in the optimal form for bioavailability. Loss of vitamin C in ironufortified cereals does not continue to increase following the initial rate increase relative to the non—fortified product. Characteristic differences between products, aside from the form of iron present, may have contributed to the distinction regarding the rates of ascorbic acid oxidation. The two products differed in initial concen— tration of iron and ascorbic acid, in potential for dissolved oxygen, in moisture content, and in pH. Additionally, the wet cereal product was thermally processed with both iron and ascorbate present, whereas mixed cereal did not receive similar heat treatment. Previous researchers have shown that the iron profile is altered by heat processing, and ferrous sulfate originally present in wet cereal may have been substan— tially converted to an insoluble, and less reactive, form. The presence of ascorbic acid with iron is reported to enhance the absorption of iron. Results reported herein indicate that loss of ascorbic acid, promoted by iron, did not occur for several days of refrigerated storage. The catalytic effect of iron is only significant in the mixed cereal product that received no heat treatment. Since it is is most likely that infant foods would be fed with only a day or two of storage, it seems that the imapct on vitamin C nutriture of infants fed vitamin Cvenriched and ironefortified foods is of minimal concern. Both_mixed cereal and wet cereal products may appropriately serve as vehicle for both vitamin Ceenrichment and for ironefortification. APPENDICES 112 APPENDIX I Split—plot Analysis Total Ascorbic Acid source of variation F sign. of F day 3.272 .05 cereal type (CT) 43.620 .001 iron 1.161 NS day X.CT 9.539 .001 day X iron 0.536 NS CT X iron 8.490 .001 day X CT X iron .0.730. NS ssE1 - 54.986 ssE2 - 37.298 MSE - 53E /8 - 6.873 1 l SSE /72 8 0.518 2 2 MS (used to test all not involving days) (used to test all involving days) 113 APPENDIX II Spliteplot analysis Reduced ascorbic acid source of variation F sign. of F day 12.311 .001 cereal type (CT) 43.259 .001 iron 0.028 NS day X CT 12.762 .001 day X iron 2.938 NS CT X iron 8.867 .001 day X CT X iron 1.119 NS SS - 63.264 SSE2 - 27.792 MSE - SSE l8 - 7.908 (used to test all not involving days) 1 l MSE . SSE I72 - 0.386 (used to test all involving days) 2 2 Split-plot Analysis 114' APPENDIX III Dehydroascorbic acid source of variation 'F sign. of P day 80.640 .001 cereal type (CT) 67.058 .001 iron 17.777 .001 day X CT 7.058 .005 day x iron 15 .511 . 001 CT X iron 1.728 NS day X CT X iron 3.122 .05 SS SS MS MS I-501.056 l I 10.041 2 I SSE /8 I 0.923 1 l I SSE I72 I 0.139 2 2 (used to test all not involving days) (used to test all involving days) 115 APPENDIX IV Bonferroni—t Analysis and Student's t Test “'k'con8tants Dehydroascorbic Acid contrast* tB sigg. of ti t sign. of t A. mixed cereal/ no iron vs w/ iron 3. wet cereal/ no iron vs w/iron C. mixed cereal vs wet cereal .8926 NS - -— (no iron) D. mixed cereal vs we. cereal .4441 NS -— - (w/iron) *Contrast A denotes the effect of iron in mixed cereal. Contrast B denotes the effect of iron in wet cereal. Contrast C denotes the difference between cereal products without added iron. Contrast D denotes the difference between cereal products with iron present. 116 APPENDIX ‘7 Split:plot Analysis Diketogulonic acid source of variation F sign. of F day ‘ 72.00 .001 cereal type (CT) 3.66 .05 iron 205.89 .001 day X CT 1.29 NS day X iron 1.39 NS CT X iron 0.54 NS day X CT X iron 1.06 NS 14.19 U) U) I 38.64 U) U) I K to I SSE /8 I 1.77 (used to test all not involving days) 1 , l SSE I72 I 0.54 (used to test all involving days) 2 2 BE 117 APPENDIX VI Bonferroni-t Analysis and Student's t Test “ k constants Diketogulonic Acid contrast* t8 ' sign. °f££~ t sign. of t A, mixed cereal] no iron vs mixed cereal w/iron .3722 NS .3722 NS B. wet cereal/ no iron vs -.9626 wet cereal w/iron us , -.9626 NS C. mixed cereal vs wet cereal .7033 NS -- - (no iron) D. mixed cereal vs wet cereal 6.1430 .01. -- -- (w/iron) *Contrast A denotes the effect of iron in mixed cereal. Contrast B denotes the effect of iron in wet cereal. Contrast C denotes the difference between cereal products without added iron. Contrast D denotes the difference between cereal products with iron present. 118 APPENDIZUVII Student's t-test distribution - evaluation of changg_over time Iron Concentration Product Day 0 Day 10 t s sign. of t (ms/1003) (111871003) Mixed cereal 1.03 0.67 2.99 2 .210 .05 (no iron) ' Mixed V a cereal 3.62 2.64 58.98 2 .029 .001 (w/iron) wet cereal 0.97 1.22 -.93 2 .465 NS (no iron) ' wet cereal 6.72 5.94 7.08 2 .190 .01 (w/iron) LIST OF REFERENCES LIST OF REFERENCES Abraham, 8., F.W. Lowenstein and C.L. Johnson. (1974) Preliminary Findings of the First Health and Nutrition Examination Survey, United States, 1971 to 1972. Rockville, Md., Office of Informs ation, National Center for Health Statistics, p. 11. Adebonojo, F.O. (1972) Artificial vs. breast—feeding: relation to infant health in a middle—class American community. Clin. Pediatr. 11:25. Ahn, C.H. and W.C. MacLean. (1980) Growth of the exclusively breast- fed infant. Am. J. Clin. Nutr. 33:183. Amine, E.K., R. Neff and D.M. Hegsted. (1972) Biological estimation of available iron using chicks or rats. J. Agr. Food Chem. 20:246. Andelman, M.B. and B.R. Sered. (1966) Utilization of dietary iron by term infants. Am. J. Dis. Child. 45:111. Anderson, H.T. and H. Barkue. (1970) Iron deficiency and muscular work performance. An evaluation of the cardio—respiratory function of iron deficient subjects with and without anemia. Scand. J. Clin. Lab. Invest. Suppl.pll4. A.0.A.C. (1970) Official Methods of Analysis. 11th ed. Association of Official Analytical Chemists. Wash., D.C., p. 777. A.0.A.C. (1975) Official Methods of Analysis. 12th ed. Association of Official Analytical Chemists. Wash. , D.C., p. 829.‘ Apte, S.V. and P.S. Venkatachalam. (1965) The effect of ascorbic acid on the absorption of iron. Ind. J. Med. Res. 53:1084. Arbeter, A., L. Echeverri, D. Franco, D. Manson, H. Valez and J.J. Vitale. (1971) Nutrition and infection. Fed. Proc. 30:1421. Bain, K. (1948) The incidence of breastefeeding in hospitals in the United States. Pediatr. 2:311. Bessey, 0.A. (1938) A.method for the determination of small amounts of ascorbic acid and dehydroascorbic acid in turbid and colored solutions in the presence of other reducing substances. J. Biol. Chem. 126:771. Bessey, 0.A. and C.G. King. (1933) The distribution of vitamin C in 119 120 plant and animal tissues,.and its determination. J. Biol. Chem. 1033687. Bjorn-Rasmussen, E. and L. Hallberg. (1974) Iron absorption from maize .. effect of ascorbic acid on iron absorption from maize supplemented with ferrous sulphate. Nutr. Metab. 16:94. Bjorn-Rasmussen, E., L. Hallberg, B. Isakssm and B. Arvidson. (1974) Food iron absorption in man. Applications of the two—pool extrinsic tag method to measure heme and nonvheme iron absorp- tion from the whole diet. J. ChhiInvest. 53:247. Bland, M.N., B.J. Constable, L.J. Harris and R.E. Hughes.(1%8) Proc. 12th Int. Congr. Chem., Food and Nutrition Section, New York, p. 155. Blaug, S.M. and B. Hajratwala. (1972) Kinetics of aerobic oxidation of ascorbic acid. J. Phar. Sci. 61 (4):556. Bothwell, T.H., G. PirziovBiroli and C.A. Finch. (1958) Iron absorp— tion. I. Factors influencing absorption. J. Lab. Clin. Med. 51:24. -- Brise, H. and L. Hallberg. (1962) Effect of ascorbic acid on iron absorption. Acta. Med. Scand. Suppl. 376:51. Brown, R.E. (1973) Breastefeeding in modern times. Am. J. Clin. NUtrg 26 : 556 q Burks, J.M., M.A. Siimes, W.C. Mentzer and P.R. Dallman (1976) Iron deficiency in an Eskimo village: The value of serum ferritin in assessing iron nutrition before and after a three- month period of iron supplementation. J. Pediatr. 88:224. Callender, S.T. and G.T. Warner. (1968) Iron absorption from bread. Am. J. Clin. Nutr. 21:1170. C.F.R. (1981) 21 Code of Federal Regulations. 100.120, 101.9. Chandra, R.K. (1973) Reduced bactericidal capacity of polymorphs in iron deficiency. Archiv. Dis. Child. 48:864. Col, J.E. (1972) Written communication to the Council on Foods and Nutrition Laboratory, Denver, Feb. 11. Committee on Nutrition. American Academy of Pediatrics. (1976) Commentary on breastnfeeding and infant formulas. including proposal standards for formulas. Pediatr. 57:278. Committee on Nutrition, American Academy of Pediatrics. (1969) Iron balance and requirements in infancy. Pediatr. 43:134. Committee on Nutrition, American Academy of Pediatrics. (1971) Ironefortified formulas,Pediatr. 47:786. 121 Committee on Nutrition. American Academy of Pediatrics. (1976) Iron supplementation for infants. Pediatr. 58:765. Conrad, M.B. and 8.6. Schade. (1968) Ascorbic acid chelates in iron absorption. A role for hydrochloric acid and bile. Gastroenter. 55:35. Cook, J.D. and E.R. Mbnsen. (1976) Food iron absorption in human subjects. III. Comparison of the effects of animal protein on non-heme iron absorption. Am. J. Clin. Nutr. 28:859. Cook, J.D., M. Layrisse, C. Martinez-Torres, R. walker, E. Monsen and 0.A. Finch. (1972) Food iron absorption by an extrinsic tag. J. Clin. Invest..51:805. Council on Foods and Nutrition. American.Medica1 Association. (1972) Iron in enriched wheat flour, farina, bread, buns and rolls. J.A.M.A. 220:855. Council on Foods and Nutrition. American Medical Association. (1941) Nutritionally improved or enriched flour and bread. J.A.M.A. 116:2849. . Crosby, W.H. (1974) Bureaucratic clout and a parable. Commentary. J.A.M.A. 228:1651. Dairy Council. (1979) Current concepts in infant nutrition. In: Contemporary Nutrition Controversies, ed. by T.P. Labuza and A.E. Sloan. St Paul: Net Publish. Comp., p. 279. Dallman, P.R. (1974) Tissue effects of iron deficiency. In: Iron in Biochemistry and Medicine, ed. by A. Jacobs and M. WOrwood. London: Academic Press, p. 437. Dallman, P.R., E. Beutler and C.A. Finch. (1978) Annotation._Effects of iron deficiency exclusive of anemia. Brit. J. Haematol. 40:179. Dallman, P.R., M.A. Siimes and E.C. Maines. (1975) Brain iron: persistent deficiency followong short-term deprivation in the young rat. Brit. J. Haematol. 31:209. Dallman, P.R.,‘M.A. Siimes and A. Slekel. (1980) Iron deficiency in infancy and childhood. Am. J. Clin. Nutr. 33:86. Danneker, D. (1966) Anemia in selected Allegheny County Health Conference populations. Pittsburgh: Allegheny County Health Dept. Dennison, D.B. (1978) Degradation of ascorbic acid in a dehydrated model food system during storage. PhD. dissertation. Michigan State University. 122 Deutsch, M.C. and C. weeks. (1965) Microfluorometric assay for vitamin C. J.A.O.A.C. 48:1248. Diamond, L.K. (1971) Iron deficiency and iron-fortified milk. Pediatr. 48:666. Dietrich, W.C., F.E. Lindquist, J.C. Miers, G.S. Bohart, B.J. Neumann and W.R. Talbert. (1957) The time-temperature tolerance of frozen foods. 4. Objective tests to measure adverse changes in frozen vegetables. Food Technol. 11:109. Elwood, P.C., D. Newton, J.D. Eakins and R.B. walker. (1968) Absorp— tion of iron from bread. Am. J. Clin. Nutr. 21:1162. Fairbanks, V.F., J.L. Fahey and E. Beutler. (1971) Clinical Disorders of Iron Metabolism, 2nd ed. New York: Grune and Stratton, p. 229. -Federal Register. (1971) Wheat flour and related products and baked products. 220:855. Filer, L.J., Jr. (1971a) Reply. Pediatr. 48:1001. Filer, L.J., Jr. (1971b) Reply. Pediatr. 48:1002. Filer, L.J., Jr. (1969) The U.S.A. today -- Is it free of public health nutrition problems? Fm Anemia. Am. J. Public Health 59:327. Finch, 0.A. and E.R. Monsen. (1972) Iron nutrition and the fortifica- tion of food with iron. J.A.M.A. 219:1462. Finch, C.A., L.R. Miller, A.R. Indamar. (1976) Iron deficiency in the rat. Physiological and biochemical studies of muscle dysfunction. J. Clin. Invest. 58:447. Finholt, P., R.B. Paulssen and T. Higuchi. (1963) Rate of anaerobic degradation of ascorbic acid in aqueous solution. J. Pharm. Sci. 52:948. 'Fisher, S.E. and A.C. Allen. (1971) Iron-fortified formulas, cost analysis. Pediatr. 48:1001. Folin, O. and H. Malnros. (1929) Blood sugar and fermentable blood sugar as determined by different methods. J. Biol. Chem. 83: .121. Pomon, S.J. (1974a) Infant Nutrition. 2nd ed. Philadelphia: W.B. Saunders Comp., p.298. Fomon, S.J. (1974b) Infant Nutrition. 2nd ed. Philadelphia: W.B. Saunders Comp., p. 359. Famon, S.J. (1974c) Infant Nutttion. 2nd ed. Philadelphia: W.B Saunders Comp., p.365. ' 123 Fomon, S.J. (1970) Public Health Serv. Publ. No. 2085, U.S. Dept. H.E.W. Food and Nutrtion Board, National Research Council. (1974) Recommended DietanrAllowances. Wash., D.C. National Academy of Sciences. Food and Nutrition Board, National Research Council. (1979) Research needs for establishing guidelines for the U.S. population. Nash., D.C.: National Academy of Sciences. Freed,iM. (1966) Association of Vitamin Chemists. Methods of Vitamin Aasay. New York: Interscience Publish., p.328. Fritz, J.C. and G.W. Fla. (1972) Vitamins and other nutrients. Applica- tion of the animal hemoglobin repletion test to measurement of iron availability in foods. J.A.O.A.C. 55:1128. Fritz, J.C., G.W. Pla, B.N. Harrison and G.A. Clark. (1974) Collabora- tive study of the rat hemoglobin repletion test for bioavailability of iron. J.A.O.A.C. 57:513. Fritz, J.C., G.W. Pla, T. Roberts, J;W. Bochne and E.L.Hose. (1970) Biological availability in animals of iron from common dietary sources. J. Agr. Food Chem. 18:647. Feurth, J.H. (1972) Reply. Pediatr. 80:526. Gardener, G.W., V.R. Edgerton, R.J. Bernard and E.M. Bernauer. (1975) Cardio-respiratory, haematological and physical performance responses of anemic subjects to iron treatment. Am. J. Clin. Nutr. 28:982. ' Gerber Products Company. (1977) Nutrient values for Gerber baby foods. Professional Communications Dept., Fremont, MI. Gerrand, J.W; (1974) Breast-feeding: second thoughts. Pediatr. 54:757. Ghosh, 8., S. Daga, D. Kasthuri, R.C. Misra and H.K. Chuttani. (1972) Gastrointestinal function in iron deficiency states in children. Am. J. Diseas. Child. 123:14. Gill, J.L. (1978) Design and Analysis of Experiments. Vol. 2. Ames, Iowa: The Iowa State University Press, p. 169. Goldsmith, G.A. (1973) Fortification of flour and bread with iron. J.A.M.A. 224:440. Gould, 8.8. and H. Schwachman. (1943) A new method for the bioassay of antiscorbutic substances. J. Biol. Chem. 151:439. Gutherie, H.A. (1975) Introductory Nutrition. St.Louis: C.V. Mosby, Inc., p. 166. 124 Haddock. S.T. and N.A.Harvey. (19711 Ironwfortified formulas. Pediatr. 48:158. ' Haddy, T.B., C. Jurkowski, H. Brody, B.J. Kallen and D.M. Czajkee Narins. (1974) Iron deficiency with and without anemia in infants and children. Am. J. Diseas. Child. 128:787. Haughton, J.C. (1963) Nutritional anemia of infancy and childhood. Am. J. Public Health 53:1121. Hegenauer, J., P.8a1tman and D.Ludwig. (1979) Effects of supplemental iron and cepper on lipid oxidation in milk. 2. Comparison of metal complexes in heated and pasteurized milk. J. Agric. Food Chem. 27:868. Herbert, V., G.W. Gottlieb, K.S. Lau, M. Fisher, N.R. Gervirtz and L.R. Nasserman. (1966) Coated charcoal assay of unsaturated iron binding capacity. J. Lab. Clin. Med. 67:855. Hill, F.L. C1967) Infant feeding: historical and current. Pediatr. Clin. N. Am. 14:255. Hillman, R.A. and C.A. Finch. (1974) Red Cell Manual. 4th ed. Phila- delphia: Davis Comp., p. 12. Hodson, A.Z. (1970) Conversion of ferrous to ferric iron in weight control dietaries. J. Agric. Food Chem. 18:946. Hoglund, S. and P. Reizenstein. (1969) Studies in iron absorption. V. Effect of gastrointestinal factors on iron absorption. Blood 34: 496. Hutcheson, R.J. (1968) Iron deficiency anemia in Tennessee among rural poor children. Fed. Health Dept. U.S. 83:939. Jackson, R.L., R. Westerfeld, M.A. Flynn, E.R. Kimball and R.B. Lewis. (1964) Growth.of “welleborn"'American infants fed human and cow‘s milk. Pediatr. 33:642. Jacobs, A. (1969) Annotation. Tissue changes in iron deficiency. Brit. J. Haematol. 16:1. Jelliffe, D.B. (1975) Unique properties of human milk. J. Reprod. ' Med. 14:133. Jelliffe, D.B. (1976) World trends in infant feeding. Am. J. Clin. Nutr. 29:1227. Jelliffe, D.B. and E.F.P. Jelliffe. (1978) Human milk in the modern world. New York: Oxford University Press. Joslyn, M.A. and J. Miller..(l949) Effect of sugars on oxidation of ascorbic acid. 1. Kinetics of autoonidation of ascorbic acid. Food Res. 14:325. 125 Joynson, D.B.M., D.M.‘Walker, A. Jacobs and A.E. Dolby. (1972) Defect of cell-mediated immunity in patients with iron deficiency anemia. Lancet (ii):1058. Karel, M. and J.T.R. Nickerson. (1964) Effects of relative humidity, air and vacuum.on browning of dehydrated orange juice. Food Technol. 18:104. Khan, M., T. Taqui and A.E. Martell. (1967) Metal ion and metal chelate catalyzed oxidation of ascorbic acid by molecular oxygen. I. Cupric and ferric ion catalyzed oxidation. J. Am. Chem. Soc. 89:4176. Kimber, C. and L.R. weintraub. (1968)‘Ma1absorption of iron secondary to iron deficiency. New Eng. J. Med. 279:453. 0 Kirk, J.R. and N. Ting. (1975) Fluorometric assay for total vitamin C using continuous flowlanalysis. J. Food Sci. 40:463. Kripke,.S.S. and E. Sanders. (1970) Prevalence of iron deficiency anemia among infants and young children seen at rural ambulatory clinics. Am. J. Clin. Nutr. 23:716. Kuhn, I.N., M. Layrisse, M. Roche, C. Martinez and R.B. walker. (1968) Observation on the mechanism of iron absorption. Am. J. Clin. Nutr. 21:1184. Labuza, T.P. (1971) Analysis of storage stability of intermediate moisture foods. Contract N.A.S. 9-12560, N.A.S.A., Houston, TX. Labuza, T.P., L. McNally, D. Gallagher, J. Hawkes and F. Hurtado. (1972) Stability of intermediate moisture foods. 1. Lipid oxidation. J. Food Sci. 37:154. Latham. M.C. (1979) International Perspectives on weaning Foods. The economic and other implications of bottle-feeding and the use of manUfactured weaning foods. In: Breast-feeding and Food Policy in a Hungry~Wbr1d. Neerork: Academic Press. Layrisse, M. and C.‘MartinezaTorres. (1971) Progress in Haematology ’VII. Ed. by E.B. Brown and C.V. Moore. New York: Grune and Stratton, p. 137. Layrisse; M,, C.~ MartineZeTorres and M. Roche. (1968) Effect of inter- action of various foods on iron absorption. Am. J. Clin. Nutr. 21:1175. Lee, K. and'F.M. Clydsdale. (1980a) Chemical changes of iron in food and drying processes. J. Food Sci. 45:711. Lee, K. and F.M. Clydsdale. (1980b) submitted publ. , 1980. 126 Lee, K. and F.M. Clydsiale. (197%() Iron sources used in food fortifica- tion and their changes due to food processing. CRC Critical Rev. Food Sci. Nutr. 11 (2):ll7. Lee, K. and F.M. Clydsdale. (1979:) Quantitative determination of the elemental, ferrous, ferric, soluble and complexed iron in foods. J. Food Sci. 44:549. Lee, K. and F.M. Clydsdale. (1981) Effect of thermal processing on endogenous and added iron in canned spinach. J. Food Sci. 46:1064. Lee, 8.3. and T.P Labuza. (1975) Destruction of ascorbic acid as a function of water activity. J. Food Sci. 40:370. Lee, Y.C., J.R. Kirk, C.L. Bedford and D.R. Heldman. (1977) Kinetics and computer simulation of as¢orbic acid stability of tomato juice as function of temperature, pH and metal catalyst. J. Food Sci. 42:640. Lockhead, A.C. and A. Goldberg. (1959) Transfer of iron to protopor- phyrin for heme biosynthesis. Role of ascorbic acid and glutathione. Lancet 2:271. Longen, K. and D. Murphy. (1978) What about W1C? National Food Review. Economics, Statistics and Cooperative Service, U.S.D.A. Sept. p. 580 Lopez, R. (1972) Incidence of anemia in full-term infants. J. Pediatr. 80:525. Lundstrom, V. and M.A. Siimes. (1977) At what age does iron supplementa- tion become necessary in lowhbirth-weight infants? J. Pediatr. 91:878. Mack, T.E., D.R. Heldman and R.P. Singh. (1976) Kinetics of oxygen uptake in liquid foods. J. Food Sci. 41:309. MBCdougall,_L.G., R. Anderson, G.M. McNab and J. Katz. (1975) The immune response in iron-deficient children: impaired cellular defense mechanisms with altered humoral components. J. Pediatr. 86:833. Macolm. E.W., J.W. Green and R.A. Swenson. (1974) J. Chem. SOC-, P- 4669. Mata, L.J., J.J. Orrutia and A. Lechtig. (1971) Infection and nutrition of children of a low socio-economic rural community. Am. J. Clin. Nutr. 29:249. .- Matthews, R.P. and J.H. Hall. (1978) Ascorbic acid, dehydroascorbic acid and diketogulonic acid in frozen green peppers. J. Food Sci. 43:532. 127 Mazor, A. (1961) Role of ascorbic acid in the incorporation of plasma iron into ferritin. Ann. New~York Acad. Sci. 92:223. Mazor, A. S. Green and A. Carleton. (1960) Mechanism of plasma iron incorporation into hepatic ferrithi J. Biol. Chem. 235:595. McMillan, J.A., S.A. Landaw‘and F.A. Oski. (1976) Iron deficiency in breast-fed infants and the availability of iron from human milk. Pediatr. 58:686. Melhorn, D.K., 8. Gross and A.J. Newman. (1971) Iron-fortified formulas. Pediatr. 48:1001. Mellader, 0., B. valquist and T. Mellbin. (1959) Breast-feeding and artificial feeding. A clinical, serological and biochemical study in 402 infants with a survey of the literature. The Norbotten study. Acta. Paediat. 48:1. Mills, M.B. and J.H. Roe. (1947) A critical study of proposed modifi- cations of the Roe and Keuther method for the determination of ascorbic acid, with further contributions to the chemistry of this procedure. J. Biol. Chem. 170:159. ascorbic acid and diketogulonic acid. Anal. Chem. 21:707. Monckeberg, B.F. (1973) Infant feeding and weaning practice: the problem as it exists in developing countries. Australian Pediatr. J. 9:48. Monsen, E.R., L. Hallberg, M. Layrisse, D.M. Hegsted, J.D. Cook, W. _ Hertz and G.A. Finch. (1978) Estimation of availability of dietary iron. Am. J. Clin. Nutr. 31:134. Morley, D. (1976) The design and use of weight charts in surveillance of the individual. In: Nutrition in Preventive Medicine, ed. by G.H. Beaton and J.M. Bengoa. world Health Organization Monogr. Series 62:520. Motzok, 1., M.D. Pennell, M.I. Davies and H.U. Ross. (1975) Effect of particle size on the availability of reduced iron. J.A.O.A.C. 58:99. Motzok, 1., R.B.. Verma, s.s. Chen, J. Rasper, R.GV.V. Hancock and H.U. Ross. (1978) Bioavailability ig_vitro solubility, and physical and chemical properties of elemental iron powders. J.A.O.A.C. 61:887. Murthy, G.K. and U.S. Rhea. (1971) Cadmium, copper, iron, lead, manganese and zinc in evaporated milk, infant products, and human milk. J. Dairy Sci. 54:1001. Nojeim, S.J. and F4M. Clydsdale. (1981) Effect of pH and ascorbic acid on iron valence in model systems in foods. J. Food Sci. 46 (2):606. 128 National Academy of Sciences. (1973) Supplementation of Human Diets with Vitamin E. National Academy of Sciences, N.R.C., Food and Nutrition Board, Nash., D.C. ' Nutrition Foundation. (1977) Guidelines for the eradication of iron deficiency anemia. A report of the International Nutritional Anemia Consultative Group, wash., D.C. Oseid, B.J. (1975) Breast-feeding and infant health. J. Clin. Obstet. Gyne. 18:149. Oski, F.A. (1972) Iron-fortified formulas in infancy: Best form of iron supplementation? J. Pediatr. 80:524. Ounsted, M. and G. Sleigh. (1975) The infant's self-regulation of food intake and weight gain. Lancet 2:1393. Qwen,G.M., A.H. Lubin and P.J. Garry. (1971) Preschool children in the United States: Who has iron deficiency? J. Pediatr. 79:563. Owen, G.M., C.E. Nelson and P.J. Garry. (1970) Nutritional status of preschool children: Hemoglobin, hematocrit, and plasma iron values. J. Pediatr. 76:761. Pearson, H.A. (1971) Commentary: Iron-fortified formulas in infancy.. J. Pediatr. 79:557. Pearson, H.A. (1972) Reply. J. Pediatr. 80:524.. Pearson, H.A. and J.H. Robinson. (1976) The role of iron in host resist- ence. Adv. Pediatr. 23:1. Pelletier, 0. (1968) Determination of vitamin C in serum. urine and other biological materials. J. Lab. Clin. Med. 72:674. Pennell, M.D., W.D. weins, J. Rasper, I. Motzok and H.U. Ross.(1975) Factors affecting the relative biobgical value of food grade elemental iron powders for rats and humans. J. Food Sci. 40:879. Penney, J.R. and 8.8. Zilva. (1945) Interfering substances in the Roe ~ and Keuther method for the determination of ascorbic acid. Biochem. J. 39:392. Pigman, W3W. and M.L. Wblfrom. (1949) Advances in Carbohydrate Chemistry, Vol. 4. New York: Academic Press, Inc., p. 189. Pla, G.W. and J.C. Fritz. (1971) Collaborative study of the hemoglobin repletion test in chicks and rats for measuring availability of iron. J.A.O.A.C. 54:13. Pla, G.W., J.C. Fritz and C.L. Rollinson. (1976) Relationship between the biological availability and solubility rate of reduced iron. J.A.O.A.C. 59:582. Pla, G.W., B.N. Harrison and J.C. Fritz. (1973) Bioavailability of reduced iron in enriched bread. J.A.O.A.C. 56:1369. 129 Plank, S.J. and M.L. Milaneski. (1973) Infant feeding and infant mortal— ity in rural Chile. Wbrld Health Organization 48:203. Purvis, G.A. (1980) In personal communication. July. Fremont, MI. Ranhotra, G.S., F.N. Hepburn and W.B. Bradley. (1971) Availability of iron in enriched bread. Cereal Chem. 48:377. Rios, E., R.B. Hunter, J.D. Cook, N.J. Smith and G.A. Finch. (1975) The absorption of iron as supplements in infant cereal and infant formula. Pediatr. 55:686. Roe, J.H..(l96la) Appraisal of methods for the determination of L- ascorbic acid. Ann. New York Acad. Sci. 92:277. Roe, J.H. (1961b) Comparative analysis for ascorbic acid by the 2, 4- dinitrophenylhydrazine method with the coupling reaction at diff- erent temperatures: a procedure for determining specificity. J. Biol. Chem. 236: 1611. Roe, J.H. (1954) Chemical determination of ascorbic acid, dehydroascorbic acid and diketogulonic acid. In: Methods of Biochemical Analysis, Vbl. I. New York: Interscience Publ., Inc., p. 115. Roe, J.H. and M.J. Oesterling. (1944) The determination of dehydroascorbic acid and ascorbic acid in plant tissues by the 2, 4-dinitrophenyl- hydrazine method. J. Biol. Chem. 152:511. Roe, J.H., M.B. Mills, M.J. Oesterling and C.M. Damron. (1948) The deter- mination of diketo-l-gulonic, dehydro-l-ascorbic acid, and 1-sscorbic acid in the same tissue extract by the 2, 4-dinitrophenylhydrazine method. J. Biol. Chem. 174:201. Saarinen, U.M. (1978) Need for iron supplementation in infants on pro- longed breast-feeding. J. Pediatr. 93:177. Saarinen, U.M., N.A. Siimes and P.R. Dallman. (1977) Iron absorption in infants: High bioavailability of breast milk iron as indicated by the extrinsic tag method of iron absorption and by the concentra- tion of serum ferritin. J. Pediatr. 91:36. Sayers, M.B., S.R. Lynch, P. Jacobs, R.W. Charlton, T.H. Bothwell, R.B. ' Welker and F.~Mayett. (1973) The effects of ascorbic acid supplement- ation on the absorption of iron in maize, wheat, soya. Brit. J. Haemat. 24:209. Sayers,'M.H., S.R. Lynch, R.W. Charlton, T.H. Bothwell, R.B. Walker and P.‘Meyett. (1974a) Iron absorption from rice meals cooked with fortified salt containing ferrous sulfate and ascorbic acid. Brit. J. Nutr. 31:367. Sayers, M.B., S.R. Lynch, R.W. Charlton, T.H. Bothwell, R.B. Walker and F. Mayett. (1974b) The fortification of common salt with ascorbic acid and iron. Brit. J. Haemat. 28:483. Schafer, W.B., WKA. Riggs and L.A, Coffin, III. (1971) Correspondence on iron—fortified formulas. Pediatr. 48:152. 130 Schaffert, R.R. and G.R. Kingsley. (1955) A rapid, simple method for the determination of reduced, dehydro—, and total ascorbic acid in biological material. J. Biol. Chem. 212:59. Schulman, I. and C.J. Smith. (1954) Studies on the anemia of prematurity. III. The mechanism of anemia. Am. J. Disess. Child. 88:582. Shah, B.G. and B. Belonje. (1973) Bioavailability of reduced iron. Nutr. Rep. Int. 7:151. Siimes, M.A., E. Vvori and P. Kutteinen. (1979) Breast milk iron -- a declining concentration during the course of lactation. Acts. Paedia. Scand. 68:29. Sjolin, S. and L. Wranne. (1968) Iron requirements in infancy and child- hood. In: Occurence, Causes and Prevention of Nutritional Anemias, ed. by G. Blix. Uppsala: Almquist and Wiksell, Symposia of Swedish Nutr. Found. VI. p. 148. Smith, N.J. and E. Rios. (1974) Iron metabolism and iron deficiency in infancy and childhood. In: Advances in Pediatrics, ed. by I. Schulman. Chicago: Year Book.Medica1 Publishers, Inc., p. 239. Srkantia, S.G., J.S. Prasad, C. Bhaskaram and K.A.V.R. Krishnamachari. (1976) Anemia and the immune response. Lancet (i):l307. Steinkamp, R., R. Dubach and C.V. Moore. (1955) Studies in iron trans- portation and metabolism. VIII. Absorption of radio-iron from.iron- enriched bread. Arch. of Inter. Med. 95:181. Stewart, R.A. (1975) Infant cereal enrichment controversy. Pediatr. 56:614. Sturgeon, P. (1959) Studies of iron requirements in infants. III. Influence of supplemental iron during normal pregnancy in mother and infant. Brit. J. Haemat. 5:45. Tannenbaum, S.R. (1976) Vitamins and minerals. In: Principles of Food Science Part 1. Food Chemistry, ed. by D.R. Fennema. New York: ’Marcel Dekker, Inc., p. 360. Ten—State Nutrition Survey in the U.S., 1968-1970: Preliminary Report to the Congress. (1971) Atlanta: Public Health Service for Disease Control. Tewari, C.P. and P.S. Krishman. (1961) Loss of ascorbic acid during estimation by the Roe-Keuther method. J. Food Sci. 26:11. Towne, J.C. and J.E. Spikner. (1963)F1uorometric microdetermination of carbohydrates. Anal. Chem. 35:211. Underwood, B.J. (1971) Iron in body tissue and fluids. In: Trace elements in human and animal nutrition, ed. by E.J. Underwood. 131 U.S.H.E.W. (1979) Healthy People. Wash., D.C.: Office of the Assist. Secr. for Health and Surgeoaneneral. Viteri, F.E. and B. Torun. (1974) Anemia and physical work capacity. Clin. Haemat. 3:609. VOjnovich, C. and V.F. Pfeifer. (1970) Stability of ascorbic acid in blends of wheat flour, CSM and infant cereals. Cereal Sci. Today 19:317. Weddell, J. (1974) The bioavailability of iron sources and their util- ization in food enrichment. Fed. Proc. 33:1779. webb, T.E. and F.A. Oski. (1974) Behavioral studies of young adolescents with iron deficiency anemia. J. Spec. Ed. 8:153. Weissberger, A., J.E. Luvalle and D.S. Thomas. (1943) Oxidation process. XVI. The autoxidation of ascorbic acid. J. Am. Chem. Soc. 65:1934. Widdowson, EIM. and C.M. Sprsg. (1951) Chemical developments in utero. Arch. Diseas. Child. 26:205. Yeomans, N.D. and D.J.B. St. John. (1975) Small intestine malabsorption of B12 in iron-deficient rats. Pathology 7:35. zdner, A.M., I. Wollinsky, G.A. Paden, J.C. Hoskin, K.C. Lewis, D.R. Lineback and R.D. McCarthy. (1980) Fortification of non-staple food items with iron. Food Technol. 34:38.