UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Antioxident effects of caseinophosphopeptides (CPP) and associated conjugates in chemical models and… Chiu, Sheila Chin Kiu 2002

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

Item Metadata

Download

Media
831-ubc_2002-0361.pdf [ 8.31MB ]
Metadata
JSON: 831-1.0090381.json
JSON-LD: 831-1.0090381-ld.json
RDF/XML (Pretty): 831-1.0090381-rdf.xml
RDF/JSON: 831-1.0090381-rdf.json
Turtle: 831-1.0090381-turtle.txt
N-Triples: 831-1.0090381-rdf-ntriples.txt
Original Record: 831-1.0090381-source.json
Full Text
831-1.0090381-fulltext.txt
Citation
831-1.0090381.ris

Full Text

A N T I O X I D A N T E F F E C T S O F C A S E I N O P H O S P H O P E P T I D E S (CPP) A N D A S S O C I A T E D C O N J U G A T E S I N C H E M I C A L M O D E L S A N D C E L L C U L T U R E S Y S T E M S I N V I T R O . B y S H E I L A C H I N K I U C H I U B . Sc., The University of Br i t i sh Co lumbia , 1999 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department o f Food Science W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A August, 2002 © Sheila Chin K i u Chiu, 2 0 0 2 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g of t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the head of my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department of food SuZftCC The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date (October 7 . <*ooz ABSTRACT Caseinophosphopeptides (CPP) derived from tryptic digests of bovine milk casein were evaluated for potential antioxidant activity. Two different commercially available (Meiji Seika Kaisha Ltd . Japan) C P P fractions, CPP-I and III, were shown to interact with ferrous iron, ferric iron and calcium ions. CPP-I and III exhibited significant (p<0.01) inhibition of both site-specific and non site-specific degradation of deoxyribose in a Fenton reaction oxidation test and protected liposomes from oxidation induced by ferrous iron and free radical A A P H . CPP-I was more effective (p<0.01) at quenching A B T S radicals than CPP-III. Laboratory synthesized C P P and CPP-saccharide conjugates were also analyzed for antioxidant activity. Different preparation techniques yield C P P of different calcium concentrations and peptide sizes. It appears that all C P P peptides carry the required polar, acidic Ser-P-Ser-P-Ser-P-Glu-Glu domain for ideal conformation for metal binding. C P P III was shown to be the most effective (p<0.01) at binding iron and lowering deoxyribose oxidation, while CPP-IJ was more effective (p<0.01) at sequestering A B T S radicals. A l l of the C P P preparations were equally effective (p>0.05) at quenching peroxyl radicals in the liposome peroxidation model. Conjugation of C P P with polysaccharide galactomannan and oligosaccharide xyloglucan did not enhance the bioactivity of C P P as predicted. Hence, both commercially available and laboratory synthesized C P P can act as a primary and also a secondary antioxidant by displaying both metal sequestering and free radical quenching activity in aqueous and emulsion models. Using a reduced serum culture system, laboratory synthesized CPP-III was also shown to be relatively non-toxic to intestinal cells and to successfully protect human intestinal embryonic Int-407 and colon carcinoma Caco-2 cells from ferrous i i sulphate-induced cytotoxicity. With a selective affinity to sequester free metal ions, such as iron, and a low toxicity in intestinal cells, bioactive casein derived peptides can act as a potential antioxidant co-nutrient that serves to remove pro-oxidative metal catalysts and free radicals from oxidizable lipids in labile food systems or biomembrane l ipid bilayer systems in vivo, especially at the site of the intestine. i i i TABLE OF CONTENTS Abstract i i Table of Contents iv List of Tables v List of Figures ix Acknowledgements x i i C H A P T E R I Literature Review 1 1.1 Introduction 1 1.2 M i l k caseins 2 1.3 Casein-derived bioactive peptides 5 1.4 Caseinophosphopeptides (CPP) 6 1.5 Properties of C P P 11 1.6 C P P and Calcium 16 1.7 C P P and Iron, Zinc and other metals 24 1.7.1 Zinc absorption 25 1.7.2 Non-heme iron absorption 25 1.8 Additional functions of C P P 31 1.8.1 Anticariogenic properties 31 1.8.2 Immunomodulating properties 31 1.9 M i l k Protein Antioxidants 33 1.10 L i p i d Oxidation and Fenton Reaction 35 1.11 Protein Glycosylation 37 C H A P T E R II Outline of Research Activties 40 2.1 General Introduction 40 2.2 Nu l l Hypothesis (H 0 ) 41 2.3 Objectives 41 2.4 Experimental Approach 41 2.5 Comments on the methodologies for antioxidant analysis 42 C H A P T E R III Antioxidant activities of C P P 47 3.1 Materials and Methods 47 3.1.1 C P P formulations 47 3.1.2 C P P Characterization studies 48 3.1.2.1 Molecular Weight Characterization . 48 3.1.2.2 Iron and Calcium Chelating Capacity48 3.1.3 Determination of Antioxidant Activi ty 49 3.1.3.1 Hydroxyl Radical Scavenging Assay in a Deoxyribose model 49 3.1.3.2 Peroxyl Radical and Metal scavenging iv assays in a Liposome model 49 3.1.3.3 Free Radical Scavenging A c t i v i t y . . . 49 3.1.4 Statistical Analysis 50 3.2 Results 51 3.2.1 Molecular Characterization of CPPms 51 3.2.2 Qualitative Metal Ion Sequestering Activity of CPPms 53 3.2.3 Site-Specific and Non Site-Specific Scavenging activity of CPPms 60 3.2.4 Effect of CPPms on the formation of Conjugated Diene 64 3.2.5 Effect of CPPms on scavenging stable D P P H and A B T S radicals 66 3.3 Discussion 69 3.4 Conclusion 76 3.5 Appendix 77 C H A P T E R IV Antioxidant activities of laboratory prepared C P P and Conjugates 82 4.1 Materials and Methods 82 4.1.1 C P P and conjugates formulations 82 4.1.2 Molecular Weight and Mineral Characterization 83 4.1.3 Hydroxyl Radical Scavenging Assay in a Deoxyribose model 84 4.1.4 Peroxyl Radical Scavenging Assays in a Liposome model 84 4.1.5 Free Radical Scavenging Assay 84 4.1.6 Statistical Analysis 85 4.2 Results 86 4.2.1 Molecular Weight and Mineral Content of C P P 86 4.2.2 Site-Specific and Non Site-Specific Scavenging activity of C P P and Conjugates. 88 4.2.3 Effect of C P P and Conjugates on the formation of Conjugated Diene 99 4.2.4 Effect of C P P on scavenging A B T S radicals. 106 4.3 Discussion 114 4.4 Conclusion 122 4.5 Appendix 123 C H A P T E R V Cytoprotective effects of C P P on colon carcinoma cells (Caco-2) and Intestinal embryonic cells (Int-407). . . . 130 5.1 Materials and Methods 130 v 5.1.1 Cel l cultures conditions 130 5.1.2 Cytotoxicity assay 130 5.1.3 Cel l Viabil i ty 131 5.1.4 Statistical Analysis 132 5.2 Results 133 5.2.1 Effect of Iron on cells 133 5.2.2 Effect of CPP-UI on cells 133 5.2.3 Effect of CPP-III on iron toxicity of cells 133 5.3 Discussion 138 5.4 Conclusion 144 General Conclusion 145 Bibliography 148 vi List of Tables Table 1. Reported sequences of C P P derived from (3-casein 8 Table 2. Reported sequences of C P P derived from ocsi-casein 9 Table 3. Reported sequences of C P P derived from a S2-casein 10 Table 4. Effect of CPPms-I and -III on % Inhibition of Deoxyribose Degradation 61 Table 5. Rate of Propagation (% Control) of Liposome Peroxidation induced by Peroxyl Radicals ( A A P H ) and % Inhibition of Lipsome Peroxidation induced by Transition Metal Ions (Fe 2 + ) 65 Table 6. Effect of CPP-I and -III on Scavenging Stable D P P H Radicals 67 Table 7. Effect of CPP-I and -III on Scavenging Stable A B T S Radicals 68 Table 8. Effect of C P P I, II and III on % Inhibition of Degradation of Deoxyribose 90 Table 9. Effect of C P P I, II and III and Conjugates on % Inhibition of Degradation of Deoxyribose 93 Table 10. Effect of C P P I, II and III and Conjugates on % Inhibition of Degradation of Deoxyribose 94 Table 11. Effect of Galactomannan and C P P I, II and III Conjugates and Mixtures on % Inhibition of Degradation of Deoxyribose 95 Table 12. Effect of Galactomannan and C P P I, II and III Conjugates and Calcium Free (CF) Conjugates on % Inhibition of Degradation of Deoxyribose 96 Table 13. Effect of Xyloglucan and C P P I, II and III Conjugates and Mixtures on % Inhibition of Degradation of Deoxyribose 97 Table 14. Effect of Xyloglucan and C P P I, II and III Conjugates and Calcium Free (CF) Conjugates on % Inhibition of Degradation of Deoxyribose. 98 Table 15. Effect of C P P I, II and III on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 100 vi i Table 16. Effect of C P P I, II and III on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 101 Table 17. Effect of Galactomannan and C P P I, II and III Conjugates and Mixtures on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 102 Table 18. Effect of Galactomannan and C P P I, II and III Conjugates and Calcium Free (CF) Conjugates on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 103 Table 19. Effect of Xyloglucan and C P P I, II and III Conjugates and Mixtures on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 104 Table 20. Effect of Xyloglucan and C P P I, II and III Conjugates and Calcium Free Conjugates on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals ( A A P H ) 105 Table 21. Effect of C P P I, II and III on scavenging stable A B T S radicals 107 Table 22. Effect of C P P I, II and III and Conjugates on scavenging stable A B T S radicals 109 Table 23. Effect of Galactomannann (G) and C P P I, II and III Conjugates and Mixtures on scavenging stable A B T S radicals 110 Table 24. Effect of Galactomannan and C P P I, II and III Conjugates and Calcium Free (CF) Conjugates on scavenging stable A B T S radicals.. . I l l Table 25. Effect of Xyloglucan (X) and C P P I, II and III Conjugates and Mixtures on scavenging stable A B T S radicals 112 Table 26. Effect of Xyloglucan and C P P I, II and III Conjugates and Calcium Free (CF) Conjugates on scavenging stable A B T S radicals 113 Table 27. Primary amino sequences of common C P P 116 Table 28. Effect of CPP-HI on the viability of Int-407 cells following a 24-hour incubation with 4 m M F e S 0 4 at 37°C 136 Table 29. Effect of CPP-III on the viability of Caco-2 cells following a 24-hour incubation with 2.5 m M F e S 0 4 at 37°C 137 Table A l . I C P - M S analysis of C P P for minerals 124 vi i i List of Figures Figure 1. Structure of the side chain of phosphoserine 3 Figure 2. Protein-polysaccharide conjugate product using Mail lard Reaction. . . . 38 Figure 3. SDS P A G E Gel (15%T) using the Laemmli buffer system (0.375 M Tris, p H 8.8). Lane 1: 40 L i g Casein, Lane 2: 40 L i g CPP-III, Lane 3: 40 L i g CPP-I , Lane 4: Molecular Weight Marker 52 Figure 4. Differential spectra of CPP-I with ferrous iron: Spectrum 1 represents the spectrum of CPP-I against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-I and F e 2 + against phosphate buffer; spectrum 3 represents the spectrum of F e 2 + against phosphate buffer 54 Figure 5. Differential spectra of CPP-I with ferric iron: Spectrum 1 represents the spectrum of CPP-I against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-I and F e 3 + against phosphate buffer; spectrum 3 represents the spectrum of F e 3 + against phosphate buffer 55 Figure 6. Differential spectra of CPP-I with calcium: Spectrum 1 represents the spectrum of CPP-I against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-I and C a 2 + against phosphate buffer; spectrum 3 represents the spectrum of C a 2 + against phosphate buffer 56 Figure 7. Differential spectra of CPP-III with ferrous iron: Spectrum 1 represents the spectrum of CPP-III against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-III and F e 2 + against phosphate buffer; spectrum 3 represents the spectrum of F e 2 + against phosphate buffer 57 Figure 8. Differential spectra of CPP-III with ferric iron: Spectrum 1 represents the spectrum of CPP-III against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-III and F e 3 + against phosphate buffer; spectrum 3 represents the spectrum of F e 3 + against phosphate buffer 58 Figure 9. Differential spectra of CPP-III with calcium: Spectrum 1 represents the spectrum of CPP-III against l O m M phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPP-III and C a 2 + against phosphate buffer; spectrum 3 represents the spectrum of C a 2 + against phosphate buffer 59 ix Figure lOa.Effect of CPP-I ( • ) and CPP-III ( • ) on Percent Inhibition of Deoxyribose Degradation (Site-Specific Assay) 62 Figure lOb.Effect of CPP-I ( • ) and CPP-III ( • ) on Percent Inhibition of Deoxyribose Degradation (Non Site-Specific Assay) 63 Figure 11. 16.5% Tris-Tricine S D S - P A G E Gel stained with Coomassie G-250 using the Laemmli buffer system (1.0 M Tris, p H 8.45) 87 Figure 12a.Effect of C P P I ( • ) , C P P II ( • ) and C P P III (LD) on Percent Inhibition of Degradation of Deoxyribose 91 Figure 12b.Effect of C P P I ( • ) , C P P II ( • ) and C P P UI (13) on Percent Inhibition of Degradation of Deoxyribose 92 Figure 13. Effect of C P P I, II and III on scavenging stable A B T S radicals 108 Figure 14. Viabil i ty of Int-407 cells following a 24-hour incubation with FeS04 a t37°C 134 Figure 15. Viabil i ty of Caco-2 cells following a 24-hour incubation with FeS04 at 37°C 135 Figure A l . Effect of C P P I at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by peroxyl radicals generated by addition of 2 m M A A P H at 37°C 77 Figure A 2 . Effect of C P P III at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by peroxyl radicals generated by addition of 2 m M A A P H at 37°C 78 Figure A3.Effect of C P P I at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50 m M F e S 0 4 at 37°C 79 Figure A 4 . Effect of C P P III at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50mM F e S 0 4 at 37°C 80 x Figure A 5 . Effect of Control ( • ) , 0 .5mM Phytic A c i d ( • ) and 0.5 m M E D T A (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50mM F e S 0 4 at 37°C 81 Figure A 6 . Effect of CPP-I at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.25 mg/ml ( X ) and I m M Phytic A c i d ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0 .50mM FeSCu at 37°C 123 Figure A 7 . Effect of CPP-JJ. at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.25 mg/ml (X ) and I m M Phytic A c i d ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0 .50mM F e S 0 4 at 37°C 124 Figure A 8 . Effect of CPP-III at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) and 0.25 mg/ml (X) and I m M Phytic A c i d ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0 .50mM FeSC>4 at 37°C 125 Figure A9.Effect of CPP-I at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) , 0.25 mg/ml (X) and 0.50 mg/ml (A) and 1.0 Ltg/ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2 m M A A P H at 37°C. . . . 126 Figure AlO.Effect of CPP-II at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) , 0.25 mg/ml (X) and 0.50 mg/ml (A) and 1.0 ug/ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2 m M A A P H at 37°C. . . . 127 Figure A l l . E f f e c t of CPP-III at 0 mg/ml ( • ) , 0.05 mg/ml ( • ) , 0.10 mg/ml ( • ) , 0.25 mg/ml (X) and 0.50 mg/ml (A) and 1.0 Lig /ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2 m M A A P H at 37°C. . . . 128 Figure A12.Preparation of C P P 129 x i Acknowledgements I would like to extend my deepest appreciation for all those who have supported me throughout my years in research. I would like to thank especially my supervisor, Dr. David Kitts, and my committee members, Dr. Li-Chan and Dr . Samuels, for their valuable advice; Dr. Soichiro Nakamura for his generous contribution of C P P samples and conjugates and also for his unconditional support; Dr. James Thompson and Joyce for their inspirations; Sherman and Valerie for their technical assistance; Hao Jing, David Popovich and Dr. Charles H u for their mentorship; and all my colleagues especially Brenda, Bonnie, Vickie , Krist i and Grace for their continual love and patience. Finally, many thanks for my parents, my brother, Chantel, Clifford, Janice and Michael for believing in me. God Bless you all! x i i C H A P T E R I: L I T E R A T U R E R E V I E W 1.1 Introduction Bovine milk and milk formula provide necessary nutritive elements, immunological agents and biologically active substances for the development of calves and human neonates (Clare and Swaisgood, 2000). M i l k proteins, consisting of 30-35g per liter of milk, are also versatile food ingredients that have functional and nutritional properties (Shah, 2000). In addition to being a good resource of high-quality proteins, milk proteins are also physiologically important by providing a range of bioactive peptides upon digestive hydrolysis (Fiat et al., 1993). These peptides have diversified functions ranging from antihypertension, immunmodulation, metal binding, opiate, antithrombotic and antibacterial activity (Schanbacher et al., 1997; Schanbacher et al., 1998; Meisel , 1998; Meisel and Schlimme, 1990; Schlimme et al, 1989). Among these peptides, caseinophosphopeptides (CPP), which are derived from the tryptic hydrolysis of casein and possessing an unique metal binding activity (Scholz-Ahrens and Schrezenmeir, 2000), warrant further attention. C P P has a high concentration of phosphorylated serine residues in close proximity on the peptide chain. A dense population of phosphate groups creates a polar, acidic domain that is electrostatically ideal for binding to divalent metals such as calcium, zinc, copper, manganese and iron (Fiat etal, 1993). In milk, the principal cation bound to the phosphate groups is calcium. Extensive research on C P P has focused mainly on the role of enhancing solubility and hence bioavailability of calcium during milk or casein meal digestion (Brommage et al, 1991; Gerber and Jost, 1986; Lee et ah, 1980; L i et al, 1989; Mykkanen and Wasserman, 1980; 1 Scholz-Ahrens et al, 1990; Yuan and Kitts, 1994; Kitts and Yuan, 1992; Tsuchita et al, 2001; Sato et al, 1983). Phosphorylated casein hydrolysis products effectively increased bone calcification in rachitic infants (Mellander, 1950) and inhibited precipitation of calcium phosphate in vitro (Reeves and Latour, 1958). Since the distal intestine is the location where the passive transport of calcium is dominant, an enhanced solubility would indirectly lead to greater calcium absorption from the intestinal lumen. Being a natural digestion product of milk (Chabance et al, 1998; Naito et al, 1974; Hirayama et al, 1992a), and resistant to further proteolytic breakdown (Kasai et al, 1992; Kasai et al, 1995), CPP carries a great potential as an in vivo bioactive peptide, especially in the distal sections of the small intestine. This affinity to form organophosphate salts with calcium and other trace elements has founded new interesting applications for CPP as dietary supplements (Steijins, 1998), antihypertension agents (FitzGerald, 1998), immunostimulants (Hata et al, 1999; Otani et al, 2000) and also anticariogenic toothpaste and mouthwash (Reynolds, 1998). CPP may also potentially aid in protecting food- or bio-systems from oxidative damage by removing transition metals responsible for catalyzing lipid peroxidation. 1.2 Milk Caseins Milk protein fractions in bovine milk include a-lactalbumin, P-lactoglobulin, caseinates, immunoglobulins, lactoferrin, proteose-peptone fractions and minor whey proteins such as transferrin and serum albumin (Clare and Swaisgood, 2000). Thus, milk proteins are mainly classified into two classes, namely caseins and whey proteins (Fiat et al, 1993). Bovine caseins are highly specific proteins synthesized in the bovine mammary gland, constitute around 78-80% of total milk proteins and are the main source 2 of amino acids, calcium and inorganic phosphate to the neonate (Schlimme and Meisel, 1995). Milk caseinates associate as highly aggregated micelles that are characterized by a high content of proline and ester-bound phosphate, but a low content of sulphur amino acids especially of cystine and low solubility at pH 4-5 (Hambraeus, 1985). Bovine casein consists of four gene products with substantially different properties: ocsi, ccS2, P and K, which represent approximately 38%, 10%, 36% and 12% of whole casein v / respectively (Fox, 2001). Caseins are mostly ^ phosphorylated and are highly diversified in the number of -CM — 0 — P —• 0~ " "2 phosphate groups attached to the protein, depending on the ^ gene product. Phosphate contents range from containing OH only 1 mole of phosphate per mole protein for K-casein, to Figure 1. Structure of the side chain of 5^  g a n ( ^ 13 phosphate groups respectively for P-casein, a si-phosphophoserine casein and ocS2-casein. Casein phosphate groups are esterified as serine monoesters (Figure 1), usually highly anionic clusters that bind bivalent cations such as calcium effectively (Fox, 2001). With a complex network of calcium phosphate bridges, bovine caseins are found to contain a content of calcium and inorganic phosphate in milk, which is far higher than would be expected from the physico-chemical solubility of calcium phosphate in milk (Hambraeus, 1985). Casein micelles are consisted of over a million submicelles enriched with K-casein on the surface and asi-, aS2- and P-caseins in the core, where the submicelles are linked through microcrystals of calcium phosphate and perhaps hydrophobic and hydrogen bonds (Fox, 2001). At a concentration over 6 mM at 30°C, excess calcium 3 binding leads to a charged neutralization of the micelle and precipitation of a s i - , aS2- and (3-caseins, while K-caseins are not affected (Fox, 2001). Precipitation and crosslinking of micelles leads to curd formation in the neonate stomach, allowing a slower release of nutrients over long time periods (Swaisgood, 1989). All caseins especially f3-casein have a high content of proline, preventing the formation of secondary structures (a helices and (3 sheets/turns), and thus allowing caseins to exist as flexible structures (Fox, 2002). Due to a flexible structure, individual caseins are especially stable to denaturing agents such as heat or urea, but are more susceptible to proteolysis than typical globular proteins through enhanced penetration by exo- and eno-peptidases (Swaisgood, 1989). A greater digestibility of casein can be associated with enhanced bioavailability of other nutrients such as trace elements, minerals and several vitamins that are bound to the protein. Thus, binding, or lack thereof, and the strength of association become the main determinants of bioavailability of the nutrient at the intestinal mucosa site (Lonnerdal, 1989). Following curding of protein in the stomach, caseins undergo extensive enzymatic hydrolysis upon entry into the small intestine and can be recovered in the jejunum mainly in the form of degraded peptides (Mahe et al., 1996). Peptides released from gastric digestion of dietary casein stimulate release of cholecystokinin (CCK) (Beucher et al., 1994) or opioid peptides (Daniel et al., 1990), slowing gastric emptying and stimulating exocrine pancreatic secretions (Mahe et al., 1996). Gastric emptying rate and gastrointestinal transit time were shown to be significantly longer with feeding of casein protein as compared to whey protein in rats, calves and humans (Daniel et al., 1990; Yvon and Pelissier, 1987; Mahe et al., 1995). Approximately half of the milk nitrogen is absorbed only between the stomach and the proximal jejunum (42%) and half (53%) 4 between the proximal jejunum and the terminal ileum (Mahe et al., 1992). This indicates that the lower part of the intestine is also necessary for the completion of the casein digestion. Through an extended absorption period, small- (i.e. d i - and tri-peptides) and medium-sized (10-51 amino acids) peptides are better absorbed intact through the intestine and wi l l potentially produce biological effects at the tissue levels (Roberts et al., 1999). 1.3 Casein-derived bioactive peptides Even though neither casein nor individual casein fractions have an established physiological role, peptides derived from casein have been shown to possess various bioactive properties (Shah, 2000). Bioactive peptides are peptides that are usually inactive within the protein sequence but become active only upon release during digestive processes (Meisel, 1998). Bioactive peptides obtained from enzymatic hydrolysis in vitro, or from digestion products in vivo, have a natural resistance to further enzymatic hydrolysis because of a high content of organically bound phosphate groups and proline residues within the peptides (Meisel and Schlimme, 1990). Sites of action for bioactive peptides are dependent on the specific sites to which they are exposed after release by proteolysis of the native protein (Schanbacher et al. 1998). Even nutritionally insignificant amounts of liberated peptides can be adequate to exert physiological effects (Meisel and Schlimme, 1990). Effects may be localized in the gastrointestinal tract after mucosal absorption. Alternatively, effects may be systemic, taking place in peripheral organs subsequent to translocation of peptides across the gut epithelium and entry as intact peptides into the blood stream (Meisel and Schlimme, 1990). 5 Bioactive peptides derived from milk caseins directly influence numerous biological processes evoking behavioral, gastrointestinal, hormonal, immunological, neurological, vasoregulatory and nutritional responses (Clare and Swaisgood, 2000). Specific functional roles include opioid agonistic, opioid antagonistic, angiotensin-converting enzyme ( A C E ) inhibitory, immunomodulatory, antimicrobial, antithrombotic and mineral binding effects. Casein-derived peptides include casomorphins (opioid agonists), casoxins (opioid antagonists), casokinins (antihypertensive peptides), casoplatelins (antithrombotic peptides), and caseinophosphopeptides (mineral carriers) (Meisel and Schlimme, 1990). Reactions associated with each bioactivity have been well characterized and reviewed (Fiat et al., 1993; Schanbacher et al., 1998; Meisel , 1997a; Meisel , 1997b; Meisel , 1998; Shah, 2000; Meisel and Schlimme, 1990; Kitts and Kwong, 2002). Two of these peptides, 6 casomorphin and caseinophosphopeptide, have been identified as digestion products found in vivo (Meisel and Frister, 1989; Chabance et al., 1998; Naito et al, 1972; Hirayama et al, 1992a). 1.4 Caseinophosphopeptides (CPP) Mineral binding bioactive peptides, caseinophosphopeptides (CPP), are classified according to the fact that they possess a dense population of phosphoseryl groups in close vicinity. The function of these peptides in bovine milk is to stabilize the emulsion by creating thermodynamically stable casein micelles that have an open sponge-like assembly for easy access of digestion enzymes (West, 1986) and can be supersaturated with protein, calcium and phosphate (Steijns, 2002). Tryptic hydrolysates of bovine casein micelles were shown to contain mineral rich peptide fractions that contain 72% of the colloidal calcium, 49% of the inorganic phosphate, 27% of the nitrogen and 82% of 6 the micellar phosphoseryl residues found originally in the native micelle (Gagnaire et al., 1996). Distribution of phosphoserine moieties varies in caseins (a s , (3 and K) and the extent of phosphorylation directly affects C P P mineral chelating affinity with the order of ocS2 > ocsi > p > K (Kitts, 1994). P-casein phosphopeptides, with molecular weight of about 3.0 kDa, are sufficient in micromolar concentration to bind and solubilize up to 40 fold molar excess of phosphate and calcium (Naito and Suzuki, 1974). Caseins are phosphorylated post-translationally as monoesters and clusters by specific protein kinases (Kinsella, 1989), following formation in the mammary gland. A particular anionic triplet of amino acid residues may act as a signal to kinase to indicate which residues to phosphorylate. Serine or threonine residues are converted to phosphorus monoesters that are located 2 amino acids on the N-terminal of negatively charged residues such as glutamic or aspartic acid (Ribadeau-Dumas, 1989; West, 1986). This anionic triplet (SerP-SerP-SerP-Glu-Glu) becomes a distinctive feature for all of the major fractions of phosphopeptides characterized, whether the C P P are of in vitro or in vivo origin. The most common C P P derived from in vitro tryptic digests of whole bovine casein include P-casein-4P (1-25), ocsi-casein-5P (59-79), ocS2-casein-4P (1-21), and ocS2-casein-4P (46-70). A wide variety of phosphopeptide sequences have been characterized using different analysis techniques or choice of hydrolytic enzymes (see Table 1, 2 and 3). Besides trypsin, protease preparations of bacterial, fungal, plant and animal origin can also be used to yield C P P (McDonagh and FitzGerald, 1998). The functional sequences of C P P required for mineral affinity have been identified to be ct si-casein (45-55), a sj-casein (56-74) aS2-casein (55-75) and P-casein (13-35) in which the anionic triplet is embedded (Kelly and McDonagh, 2000). 7 Table 1. Reported sequences of CPP derived from p-casein. Caseinate M o l Wt Peptide Enzyme Source Reference P 3.1kDa 1 , 2 ' 3 (fl-25)4P Trypsin In vitro Manson and Annan, 1971; Ono et al, 1994; Gagnaire et al, 1996; Cross et al, 2001; Kitts etal, 1991; Berrocal et al, 1989; Schlimme and Meisel , 1995 P (fl-25)4P Pepsin & Trypsin In vitro Ono etal, 1998 P 3.5 k D a 3 3.3 k D a 4 (fl-28)4P Trypsin In vitro Ono etal, 1994; Hirayama et al, 1992b; Gerber and Jost, 1986; Gagnaire et al, 1996 P (f2-28)4P Trypsin In vitro Juillerat et al, 1989 P 3.6 k D a 3 (fl-29)4P Trypsin In vitro Gagnaire et al, 1996 P S.OkDa 1 (f2-25)4P Trypsin In vitro Ono etal, 1994; Reynolds etal, 1994; Kopraera / . , 1992 P (f7-24)4P Gastric and Intestinal enzymes In vivo (rats' ileum) Hirayama et al, 1992a P (f7-24)4P Pancreatin In vitro Adamson and Reynolds, 1995; Feng etal, 1997 'Reynolds et al, 1994 2 Ono et al, 1998 3Gagnaire et al, 1996 4Gerber and Jost, 1986 8 Table 2. Reported sequences of CPP derived from ctsi-casein. Caseinate M o l Wt Peptide Enzyme Source Reference OCsl 5.6 k D a 3 (f35-79)8P Trypsin In vitro Gagnaire et al, 1996 OCsl 5.5 k D a 3 (f37-79)7P Trypsin In vitro Gagnaire et al., 1996 Osl 4.6 k D a 1 ' 3 (f43-79)7P Trypsin In vitro Reynolds et al, 1994; Gagnaire et al., 1996 Osl (f46-70)4P Trypsin In vitro Ono etal, 1994 OCsl (f59-78)5P Pancreatin In vitro Adamson and Reynolds, 1995 Osl 2 .7kDa 1 , 2 , 4 (f59-79)5P Trypsin In vitro Ono etal, 1994; Berrocal etal, 1989; Juillerat et al, 1989; Huq et al, 1995; Reynolds et al, 1994; Gerber and Jost, 1986 Osl (f59-79)5P Gastric and Intestinal enzymes In vivo (rats' ileum) Brommage et al, 1991 OCsl (f59-79)5P Pepsin & Trypsin In vitro Ono etal, 1998 OCsl (f61-74) Gastric and Intestinal enzymes In vivo (rats' ileum) Hirayama et al, 1992a OCsl (f61-78)5P Pancreatin In vitro Adamson and Reynolds, 1995; Feng etal, 1997 Osl (f62-69) Gastric and Intestinal enzymes In vivo (rats' feces) Kasai etal, 1992 Osl (f64-73) Gastric and Intestinal enzymes In vivo (rats' feces) Kasai et al, 1992 Osl (f59-64)5P Trypsin In vitro Gerber and Jost, 1986 OCsl (f66-74)3P Gastric and Intestinal enzymes In vivo (minipigs' intestine) Meisel andFrister, 1988; Meisel and Frister, 1989a; Meisel and Frister, 1989b 'Reynolds etal, 1994 2 Ono etal, 1998 3Gagnaire et al, 1996 4Gerber and Jost, 1986 9 Table 3. Reported sequences of CPP derived from otS2-casein. Caseinate M o l Wt Peptide Enzyme Source Reference ocs2 2.7 k D a1 ' 3 (fl-21)4P Trypsin In vitro Kitts etal, 1991; Reynolds etal, 1994; Gagnaire et al, 1996 a S 2 3.1 k D a3 (fl-24)4P Trypsin In vitro Gagnaire et al, 1996 CCs2 (fl-32) Trypsin In vitro Hirayama et al, 1992b 2.6 k D a 1 (f2-21)4P Trypsin In vitro Ellegardef a/., 1999; Reynolds etal, 1994; Kopraef a/., 1992 a S 2 3.0 k D a 1 , 4 (f46-70)4P Trypsin . In vitro Gerber and Jost, 1986; Ellegard et al, 1999; Juillerat et al, 1989; Reynolds etal, 1994; Kopraera / . , 1992 'Reynolds etal, 1994 2 Ono etal, 1998 3Gagnaire et al, 1996 4Gerber and Jost, 1986 Phosphopeptides prepared through different in vitro or in vivo methods were shown to share general amino acid profiles, analogous chemical natures and physiological functions (Naito and Suzuki, 1974). Several C P P derived from P-casein were shown to exist in the stomach and duodenum of adult humans after milk digestion (Chabance et al, 1998) and in intestinal digests of mini-pigs (Meisel and Frister, 1988; Hirayama et al, 1992a) and rats (Naito et al, 1972) hours following ingestion of casein diets. Because of the high concentration of negative charges on phosphopeptides, C P P becomes relatively resistant to further proteolysis (Meisel and Schlimme, 1990) and remnants of it can be found in feces of casein-fed rats, suggesting at least a part of the C P P formed in the small intestine was not hydrolyzed in the digestive tract but excreted (Kasai et al, 1992). Therefore, C P P in small intestine not only resists hydrolysis by digestive enzymes, but may even escape hydrolysis by enteric bacteria in the distal 10 intestine (Kasai et al, 1992). C P P found in the cecum and colon contents of rats fed diets containing purified C P P was not distinguishable from that of rats fed casein that contained a comparable amount of phosphoserines (Kasai et al., 1995). 1.5 Properties of CPP Fractions of phosphopeptides isolated from casein hydrolysates were found to sequester calcium phosphate at neutral to alkaline p H of 7 - 10.5 (Reeves and Latour, 1958; Swaisgood, 1982) and inhibit amorphous calcium phosphate precipitation (Sato et al, 1991; Berrocal et al, 1989; Ellegard et al, 1999). B y linking the seryl phosphate groups to calcium phosphate in spherical, nanometer-sized particles, C P P , specifically the P-CPP (1-25), is capable of stabilizing amorphous dicalcium phosphate. This activity wi l l arrest the formation of hydroxyapatite crystals at an early stage (Holt et al, 1998). Phosphoseryl groups undergo different phases in binding kinetics upon increasing concentrations of inorganic calcium. At low concentrations below 1 m M calcium, phosphoseryl residues exhibit primarily exothermic binding to calcium ions. As concentration increases to 1-3 m M , a second exothermic binding phase is followed by a progressively more endothermic reaction involving the phosphoseryl residues binding to carboxylate residues, resulting in increased self-association. Above a 3 m M calcium threshold concentration, the reaction becomes very endothermic and phosphoserine residues interact predominantly with carboxylate residues. Gradually, persistent binding to carboxylate residues minimizes inter-molecular electrostatic repulsion and hydrophobic interaction of the hydrophobic domains leading to the formation of large aggregates and eventual precipitation between 5-6 m M of calcium. (Swaisgood, 1993). 11 Inhibition of calcium phosphate precipitation can be achieved at a concentration of lOmg/L of C P P or higher at a p H 6.5 with complete emulsion stability maintained with lOOmg/L (Berrocal et al., 1989). Strength of calcium sequestering is in turn highly correlated to the extent of phosphorylation in different fractions of C P P and conditions (pH, ionic strength, temperature and enzyme selection) under which peptide-metal interactions occur. The apparent calcium binding constant, K a p p (1/mol), reported by Meisel (1997) was 328 and the lowest for a si-casein (f43-48) that consisted of 2 phosphate groups as compared to 629 for P-casein (f 1-25) and 841 for a si-casein (f59-79) with 4 and 5 phosphate groups respectively. Optimal binding of C P P to 4 5 C a occurred at p H 8.0 with a binding capacity to 4 5 C a that was determined to be 50% at 8 u,g of C P P and 100% at 25 (ig C P P (Nagasawa et al, 1991). In order to perform C P P bio-sequestering activity, phosphorylated serine clusters have a critical role to not only create an anionic environment for effective mineral binding, but also to maintain the stability of peptides towards further proteolytic hydrolysis. Peptides, in which the O-phosphorylserine groups are dephosphorylated, have been shown to be more susceptible to enzymatic hydrolysis than in cases where organic phosphate groups were still bound as monoesters, diesters or triesters (Mellander, 1963). Phosphopeptide a si-casein (59-79) obtained most abundantly in the ileum of rats following casein diets was shown to be more slowly dephosphorylated than other peptide fragments in the intestinal lumen, thereby implicating a high in vivo stability (Brommage et al, 1991). When completely dephosphorylated, C P P loses the affinity for divalent metals (Nagasawa et al., 1991) and calcium solubilization properties (Gerber and Jost, 1986). Dephosphorylated C P P was ineffective (Berrocal et al, 1989) even at 12 concentrations higher than lOOmg/L at preventing calcium phosphate precipitation, indicating that the binding sites for calcium and functional units of C P P bioactivity are essentially the phosphoserine residues. Cation binding affinity (pK) of the phosphoserines can be greatly affected by an alteration in p H , temperature, ionic strength and the presence of calcium. Baumy and colleagues (1989) identified a maximum calcium binding (pK 6.57-7.10) taking place on residue 17,18 and 19 on P-casein (1-25). In accordance to the calcium binding phases of phosphoserine residues, protonation of the phosphate groups at a low p H environment hinders calcium binding and calcium phosphate solubilization, whereas competitive binding of calcium by carboxylate groups from residual dissolved carbon dioxide under alkaline conditions also inhibits binding affinity (Meisel and Olieman, 1998). Increasing temperature from 20°C to 40°C wi l l catalyze endothermic binding of calcium, thereby greatly enhancing CPP-metal association (Swaisgood, 1993). At greater ionic strength, the presence of other cations such as sodium, with very similar binding kinetics as calcium, competes with calcium to bind phosphoseryl residues, thus lowering the calcium-binding ability of C P P . The presence of calcium is also inhibitory to calcium chelation, shown by a linear decrease in binding between 1 and 4 moles of calcium per mole of phosphopeptide. However, once the four sites on P-casein (1-25) are saturated, any excess calcium has a minor effect on binding affinity (Baumy et al., 1989). Binding of one calcium atom to ionized phosphoserine can exert an inductive effect by attracting electrons to decrease the strength between the oxygen and hydrogen bond of the second acidic group. This observation indicates that not only is the functional binding sequence (SerP-SerP-SerP-Glu-Glu) of C P P important to mineral-13 peptide interaction, but the neighboring residues are also equally essential for effective binding (West, 1986). Amino acid residues upstream and downstream from this region may also participate in binding. Evidence showed that when the sequence that spanned this domain (SerP-SerP-SerP-Glu-Glu) was mimicked synthetically, calcium binding activity was lost (Meisel and Frister, 1989a; Shah, 1990). Synthetic peptides corresponding to the phosphoserine rich region of a si-casein bind less calcium than the entire tryptic peptide (59-79), whereas the synthetic N-terminal (59-63) and the corresponding C-terminal (71-79) do not bind calcium at all (FitzGerald, 1998). Similarly when the sequence is cleaved at the glutamic acid residue, charge redistribution in the anionic cluster significantly lowers calcium-binding activity (Park and Al len , 1998). Besides physical parameters that are characteristics to C P P bioactivity, the enzyme selection for C P P preparation is also a main contributing factor to its bioactivity. Caseinophosphopeptides prepared from a selection of bacterial, fungal, plant and animal proteolytic enzymes could bind 0.40-0.61 and solubilize 7.4-24.0 mg Ca 2 + /mg C P P , as compared to traditional C P P prepared with trypsin that binds 0.60 and solubilizes 14.7 mg Ca 2 + / mg C P P (McDonagh and FitzGerald, 1998). Peptides prepared by successive digestion treatment with pepsin (pH 4.0 stomach) and trypsin (pH 7.0 intestine) were shown to have a higher capacity to bind calcium. For example, these peptides were reported to chelate 248 mg C a 2 + / g peptide, as compared to tryptic peptides that bind 168 mg C a 2 + / g peptide (Ono et al., 1998). Using an innovative in vitro simulation of human digestion by stepwise addition of digestive enzymes and alteration of p H , C P P that was generated by porcine pancreatic trypsin, bioprotease N100L and porcine pancreatin did 14 have significantly different effects on the dialysability of 4 3 C a (Kennefick and Cashman, 2000). The strength of calcium binding to peptide seems to be related primarily to the nature of peptide and preparation techniques. The stability of the metal-peptide complexes, on the other hand, is dependent on the conformational structure of the peptides and metals. Information on the actual conformation of the peptides is limited but past research based on circular dichroism and optical rotatory dispersion has shown that, when compared to other globular proteins, bovine a- and P-caseins have unusually random, open and flexible structures with little secondary structures, particularly a-helix (Swaisgood, 1993). Divergence from the behavior of typical globular proteins may result from the acidic, highly solvated domain, which allows the caseins to approach near random coil properties. Bovine a- and P-caseins are characterized by a high frequency of phosphoseryl residues and proline resides, having a distinct amphipathic nature with both polar and hydrophobic domains organized in a tertiary structure (Swaisgood, 1982). Predicted secondary structures around sites of phosphorylation in the Ca 2 +-sensitive caseins often comprise of an a-helix-loop-a-helix motif with the sites of phosphorylation in the loop region (Holt et al., 1989). The motif consisting principally of phosphoserines is essential for binding calcium, but possible conformation changes in the backbone of C P P allows for residues remote from the phosphorylated motif to also interact with calcium (Cross et al., 2001). Recent N M R studies showed that the majority of highly phosphorylated sites preferentially exist in regions likely to be P-turns or loops in the peptide structure. This 15 feature corresponds to different fractions of C P P . Characterization by N O E (nuclear Overhauser effect) connectivities and ' H - N M R showed that although different fractions of C P P share analogous sequence and functions, they have distinctly different conformations. The phosphoseryl motif in the peptide a si-casein (59-79), segment G l u 6 1 -SerP 6 7 , was implicated in a loop-type structure. This motif was followed by a P r o 7 3 - V a l 7 6 in |3-turn conformations (Huq at al, 1995), while the same motif in P-casein (1-25) S e r P 1 7 - G l u 2 0 was embedded in P turns along with sequences V a l 8 - G l u n , G l u 2 1 - T h r 2 4 and A r g ' - G l u 4 ; all of which are located in a loop structure (Cross et al, 2001). 1.6 C P P and C a l c i u m More than 50 years ago, peptides isolated directly from enzymatic hydrolysates of casein with characteristically high phosphorus content were found to form very soluble calcium complexes. These peptides, named casein phosphopeptides (CPP), were associated with enhancing calcium absorption and hence skeletal mineralization in raichitic infants (Mellander, 1950). Since the discovery of C P P , studies have consistently shown C P P to inhibit calcium phosphate precipitation in vitro (Reeves and Latour, 1958) and thus enhance solubility and uptake of calcium in vivo from the distal intestine during luminal digestion of casein-containing meals (Naito et al, 1972; Sato, 1986; Lee et al, 1992). C P P formed from digestive processes, yield approximately 5mg for 200g casein/kg diet (Lee et al, 1980) and can improve paracellular calcium absorption in situ in the distal intestine of rats fed casein diets (Nagasawa et al, 1991) and soy diets supplemented with C P P (Yuan and Kitts, 1992). In addition to alkaline conditions of the intestine, C P P was also shown to function at a low pH, gastric environment, accelerating 16 calcium emptying and lowering insoluble calcium level in rats following consumption of calcium fortified milk (Tsuchita et al., 2001). Once in the systemic circulation, C P P may have an intricate role in modulating calcium metabolism. C P P has been implicated in stimulating bone and cartilage calcification, increasing bone mass and calcium content (Lee et al., 1992) and reducing bone resorption (Ashida et al., 1996; Tsuchita et al., 1996). It was suggested that C P P has the ability to enhance bone mineralization by activating alkaline phosphatase (Matsui et al., 1994), enhancing renal tubular reabsorption of calcium and increasing serum osteocalcin (Tsuchita et al., 1995) and to lower bone resorption by lowering tartrate-resistant acid phosphatase activity, osteoclast differentiation and cell number (Matsui et al., 1994) and lowering urinary c A M P concentrations (Tsuchita et al., 1995). C P P significantly enhanced the accumulation of calcium in the tibia of chicks (Mykkanen and Wasserman, 1980) and significantly improved hypertrophic cartilage calcification in explanted embryonic rat bone rudiments at a concentration of 10 mg/lOOml of culture medium (Gerber and Jost, 1986). These observations suggest that the inhibitory effect of C P P on amorphous calcium phosphate precipitation does not exclude local precipitation and hence mineralization in the cartilaginous matrix. Approximately one third of the calcium in bovine milk is present in the serum phase as free C a 2 + or complexed by phosphate, while two thirds is partly incorporated in micellar calcium phosphate and partly bound to casein (Tsuchita et al., 2001). Intestinal calcium absorption is the sum of two processes; a saturable transfer mediated by hormone- and vitamin D-dependent active transport via calcium binding proteins in the proximal intestinal (i.e. duodenum and upper jejunum) and a non-saturable transfer that is 17 concentration-dependent, diffusional and predominantly paracellular in the lower intestine (Saito et al, 1998). In the case of intestinal absorption after a meal, the concentration of calcium in the lumen reaches between 3-4 mmol/L in rats and 7-8 mmol/L in humans (Bronner et al., 1987). Calcium flows through the apical membrane of enterocytes by active and passive mechanisms, are driven by high extracellular or intracellular concentration gradient and sustained by the calcium pump located at the basal membrane that pushes calcium into the capillary compartment (Ferraretto et al., 2001). At high intraluminal calcium levels, the proportion of calcium absorbed by diffusion becomes proportionally greater than by active transport (Saito et al., 1998). When the relationship between C P P concentration and the corresponding level of serosal calcium in chicks is mapped on a graph, the occurrence of a non-linear plot suggests that perhaps 2 different types of transport processes are involved: a saturable absorption of calcium at low peptide concentrations and a non-saturable translocation of calcium stimulated by a high concentration of peptides. The authors postulated that the peptides are capable of acting at a minimum of two sites in the distal intestine, the site of uptake and the site of release of calcium from the epithelium membrane (Mykkanen and Wasserman, 1980). Evidence of the non-saturable transfer is shown by the independence of C P P enhanced calcium transport from vitamin D status in chicks and rats (Mykkanen and Wasserman, 1980; Kopra et al, 1992). While some researchers have theorized that the physiological role of C P P on intestinal calcium absorption is only an indirect luminal inhibition of the precipitation of phosphate calcium salts (L i et al, 1989), and not a direct effect on the intestinal mucosal membrane (Erba et al, 2001), some are convinced that C P P has a much greater potency 18 as a bioactive peptide. In the presence of extracellular calcium, C P P was shown to induce a transient rise of free intracellular calcium ions in intestinal tumor cells HT-29 and Caco-2 (Ferraretto et al., 1999; Ferraretto et al., 2000). Since calcium flux generated by C P P lacks desensitization, C P P appears to aggregate in oligomers suitable for insertion into plasma membrane. It is possible that C P P does not influence membrane-bound receptors or ion channels, but acts rather as a calcium carrier for calcium-selective channels on the plasma membrane or becomes internalized as a calcium-peptide complex via endocytosis (Ferraretto et al., 2001). While many of the studies have shown that C P P has a positive effect on calcium solubility and subsequent metabolism and utilization of calcium, there exist many discrepancies of findings among different studies. For example, it is understandable that C P P functions can be influenced by variation in experimental designs, preparation methods (i.e. in vivo or extrinsic), extent of interaction between C P P and calcium and also end measures used to determine calcium absorption. Dietary calcium levels used in experimental design, particularly the calcium to C P P ratios, are important for determining the effect of C P P on calcium absorption in animal studies. Saito and colleagues (1997) have identified the minimum effective dose of C P P to enhance calcium absorption, under marginal dietary calcium levels (0-0.35%), to be 0.7g/kg or a weight ratio of C P P / C a of 0.2 in rats fed C P P supplemented soy protein diets. Similarly, C P P boosted calcium balance and plasma calcium in rats fed isolated soy protein diets in a dose-dependent manner when the C P P / C a ratios range between 0 to 0.35, reaching constant levels when C P P / C a ratios are above 0.35 (Lee et al., 1992). 19 In contrast, C P P seems to be quite ineffective in enhancing calcium absorption and utilization at higher levels of calcium intake. Soy diets and casein diets supplemented with C P P (3%) had little effect on apparent calcium absorption and utilization in rats with an adequate (0.5%) calcium intake (Yuan and Kitts, 1992). A t a level of 1.8 g/lOOg, C P P fed as a component of whey protein diet (Ca 0.7%) did not affect plasma calcium, apparent calcium absorption or retention and femur mineralization in vitamin D-replete and -deficient rats (Kopra et al, 1992). C P P at levels of 3.8 or 7.6 g/lOOg in whey protein diets (Ca 0.53%) also had no effect on fractional calcium absorption in normal female rats (Brommage et al, 1991). However, calcium absorption in the same female rats was elevated when 9-27% casein was supplemented in whey protein diets when total calcium intake was dropped to 0.2%. Inclusion of C P P in low (1.5%) calcium diets successfully improved utilization of calcium and recovered specific gravity of egg shells in laying hens, but at normal (3.4%) calcium diets, calcium requirements for eggshell formation were satisfied and C P P addition did not have any effects on egg shell parameters (Ashida et al, 1996). A t an adequate dietary calcium intake level, luminal calcium concentration may already be quite high so that the contribution of C P P to enhancement of calcium may be very small or negligible (Saito et al, 1998). It appears important that when casein is added as supplementation to animal diets, the basal diet must also include other dietary factors such as carbohydrate and fatty acids in order for effective in vivo release of C P P and its subsequent mineral-binding bioactivity (Lee et al, 1983). Besides variable dietary calcium levels in experimental designs, inconsistent C P P preparation methods and associated purity standards also contribute to the diversity of 20 conclusions from different studies. Crude preparation and demineralization of C P P have been shown to greatly lower the affinity to sequester calcium and hence exert an in vivo effect. Scholz-Ahrens and colleagues (1990) have attributed a lack of C P P effect on calcium content and bending moment of femurs from weanling piglets and on apparent absorption or retention of calcium in vitamin-D depleted rats to the low quality of phosphopeptides in casein meals. Decalcification of C P P prior to supplementation has proven to have no effect and even negative effect on calcium absorption in rats. A t a concentration of lOOg "decalcified" CPP/kg in a single meal, the absorption of calcium was not affected but was significantly reduced by inclusion of 200, 350 and 500g CPP/kg (Bennett et al., 2000). The affinity of C P P , either ingested or formed from digestive processes, to exert a function on calcium at any dietary calcium levels is dependent on the extent of interaction between C P P and the mineral. The efficiency of C P P in calcium solubilization under in vitro conditions may be compromised in the presence of phosphatases, inhibitory agents and other dietary factors, found in human and animal models. It was shown that a C P P -C a complex could not be absorbed and remained on the mucosal side of the ileum when extracted rat intestinal tissues are mounted in an Ussing chamber (L i et al., 1989). This observation is however not valid for extrapolation to in vivo studies, since C P P and C P P -Ca complexes are capable of interacting with intestinal phosphatases, thus influencing subsequent utilization of the calcium that is released. Intestinal phosphatases were found to have the highest activity in the ileum of rats (Brommage et al., 1991), breaking calcium ions free from the C P P - C a complex and releasing them for passive absorption. 21 C P P was also shown to interact with alkaline phosphatase to enhance bone mineralizaiton and with tartrate-resistant acid phosphate to reduce bone resorption (Matusi et al, 1994) The presence of inorganic phosphate or phytates in the diet wi l l lead to the formation of insoluble calcium salts, significantly affecting the extent of CPP-calcium interaction and lowering the potency of C P P as mineral chelator. A n y interference with the interaction between C P P and calcium from the digestion of other food ingredients wi l l not only render C P P ineffective in enhancing calcium absorption, but also severely hinder calcium bioavailability. In the presence of inorganic phosphate, calcium transport measured using a perfused ileum was decreased by over 90% and adding C P P to the solution fraction minimized the inhibitory effect of phosphate on calcium absorption to around 46-60% in rats (Erba et al, 2001). Using a perfused intestine technique, Hansen et al. (1996) showed a negative impact on calcium and zinc absorption in the presence of aqueous phytate solution. Although adding C P P did not completely overcome the effect of phytate, C P P was significantly better than bovine serum albumin, casein or whey proteins to inhibit the phytate effect on zinc and calcium absorption. C P P was also shown to improve calcium absorption in rat pups from oat-based and soy-based high phytate infant diets by 45 and 10%, respectively (Hansen et al., 1996). In humans, C P P was associated with an improvement of calcium and zinc absorption from phytate containing rice-based infant cereal by 26-27% and 24-32%, respectively, when the ratio of phytate to calcium or zinc was relatively low (phytate:Ca:Zn ratio (Limol) of 30:12:19.7). However, C P P did not affect calcium absorption from whole grain-based infant meals and bread meals (Hansen 22 et al, 1997a; Hansen et al., 1997b), when phytate to calcium or zinc ratios were much higher (phyate:Ca:Zn ratios (umol) of 300:13.5:27.1 and 528:7.8:21.8). Interestingly, not only is the extent of interaction between C P P and calcium influenced by other dietary factors, C P P may also be compatible with other bioactive compounds in the diet, thus enhancing calcium utilization in this way. While the isoflavone genistein was found to increase dry weight, calcium content, alkaline phosphatase activity and D N A content in femoral diaphyseal and metaphyseal tissue of young and elderly rats, supplementation of diets with C P P augmented the effect of genistein on calcium content and alkaline phosphatase activity of metaphyseal tissues in young and elderly rats in a synergistic manner (Ma and Yamaguchi, 2000). In addition to dietary factors tested, variation in the measured effect of C P P on calcium absorption may also be associated with a wide diversity in end measure parameters used in calcium absorption studies. Many authors have persistently employed similar methods such as serum/urinary/fecal calcium (Kopra et al., 1992; Lee et al., 1992; Saito et al., 1997), ligated intestinal loop (Mykkanen and Wasserman, 1980; Lee et al., 1980; Kitts et al, 1992; Yuan and Kitts, 1991; Yuan and Kitts, 1992) and bone calcification parameters, such as femur calcium content and bending moment (Pointillart and Guegen, 1989; Scholz-Ahrens et al, 1990; Yuan and Kitts, 1994), to calculate apparent absorption of calcium. Apparent absorption, based on calcium ingested and excreted via urine or feces, may not be a sensitive method to assess bioavailability of calcium, because it is affected not only by exogenous calcium sources but also by endogenous calcium excretion in feces. It is possible that the stimulative action of C P P on calcium absorption is so slight that the changes in apparent absorption of calcium and 23 bone metabolism that is controlled primarily by hormones that maintain calcium homeostasis were not detected in some cases (Matsui et al, 1994). Therefore, other authors become more innovative and make use of fecal excretion radioisotope ratios 4 7 S c : 4 7 C a (Brommage et al, 1991; Bennett et al, 2000), other calcium controlled parameters such as systolic/mean blood pressure (Nagasawa et al., 1991; Kitts et al., 1992) and alternative bone mineralization indices like alkaline phosphatase (Matusi et al., 1994), osteocalcin (Tsuchita et al, 1995) and bone D N A (Ma and Yamaguchi, 2000). However, the lack of uniformity in end-point measures makes it especially difficult to make comparisons, substantiate and determine validity of results among studies. 1.7 CPP and Iron, Zinc and other metals Phosphoserine clusters on casein-derived phosphopeptides bind calcium as well as other cations through coordinate or ionic bonding. Caseins usually bind about 95% of manganese and zinc and 50-75% of copper and iron in bovine milk (Flynn and Cashman, 1997). The specific conformations adopted by C P P are often influenced by the specific cation involved. Caseinophosphopeptides tend to configure in structures consisting of turns and loops in the presence of calcium ions but they behave markedly different in the presence of other counter-ions such as sodium or ammonia counter-ions (Cross et al, 2001). One molecule of a s ]-casein can bind 14 atoms of iron and 16 atoms of copper while |3-casein can bind 9 atoms of iron, 13 atoms of copper and 7 atoms of calcium at p H 7.0-8.0 (Brule and Fauquant, 1982; Baumy and Brule, 1988). Under physiological p H conditions, stability of manganese and zinc complexes has been shown to be greater than that for calcium and magnesium and still greater than strontium (Mellander, 1963). Calcium, magnesium, zinc and maganese are bound to phosphoserines through ionic 24 linkages, whereas copper and iron are bound through ionic and coordination links with N H 2 , C 0 2 H , C O N H and H 2 0 (Baumy and Brule, 1988). Therefore, while CPP-calcium, magnesium, zinc or manganese complexes were vulnerable to changes in p H and ionic strength, CPP-iron and copper complexes were not affected. It was found that coordination linkages developed between iron and phosphoserine were approximately 100 fold stronger than the ionic bonding between phosphoserine and zinc or calcium (Bouhallab et al, 1991; Brule and Fauquant, 1982). 1.7.1 Zinc absorption Presence of inorganic iron and phytates can hinder intestinal zinc absorption in both animals and humans. C P P acts as a low-molecular chelator similar to ligands such as histidine, E D T A and organic acids to form a complex with zinc, thereby enhancing zinc absorption. C P P was also responsible for overcoming the inhibitory effect of dietary phytates in aqeous phytate solutions (Hansen et al, 1996) or phytate-containing rice meals (Hansen et al, 1997a) and that of inorganic iron on zinc absorption (Sandstrom et al, 1985; Lonnerdal et al, 2000; Peres et al, 1998). 1.7.2 Non-heme iron absorption Casein binds about 85% of the added non-heme iron in skim bovine milk; 72% and 21% were associated with the ccs- and P-caseins (Demott and Dincer, 1976), respectively. Most of the iron added to bovine milk binds to phosphoserine residues in the casein micelle, while some are sequestered by inorganic phosphate in both the milk serum and casein micelle (Hunt, 2001). Iron bound to phosphoserine is stable to changes in p H between 2 and 6, suggesting that iron is probably not only bound to casein by ionic bonds but also by coordination bonds via the oxygen of the phosphate group in a 25 tetrahedral coordination structure (Galdi and Valencia, 1988). However, at concentrations above 4 m M , iron-casein complex precipitates through negative charge neutralization of caseins by iron and formation of intermolecular iron bridging (Gaucheron et al., 1996). Non-heme iron bioavailability is usually low, ranging between 2 to 20%, and inversely related to iron stores or influenced by enhancing and inhibiting factors present in the diet (South et al., 2000). Iron from food is released under acidic conditions in the stomach and can form large, insoluble polymers of F e + 3 O H when p H rises in the duodenum. Major pathways of non-heme iron uptake involve brush border vesicles or an energy-requiring, carrier-mediated process via a transmembrane protein transporter. However, the pathway by which dietary factors have the strongest influence on divalent mineral absorption is the paracellular route. Although this route is a minor pathway, there are no adaptive regulation mechanisms involved, allowing the accommodation of significant amounts of iron when dietary iron levels are high (Benito and Mil le r , 1998). It is believed that the ferrous form of iron, being easily diffusible, is the form primarily absorbed in humans (Peters et ah, 1971). The presence of dietary factors may serve to enhance or inhibit non-heme iron absorption. Through multiple regression analyses on various dietary factors in humans, the main biochemical predictors of non-heme iron absorption from complex meals were identified to be sources of animal tissues, phytic acid and ascorbic acid (Reddy et al., 2000). While animal tissues (South et al., 2000) and reducing agents, such as ascorbic acid (Hallberg et al., 1986; Cook and Monsen, 1977), were found to have a positive effect on iron absorption, inorganic calcium (Hallberg et al., 1991), phytate (Hallberg et 26 al, 1987) and phosphate (Cook and Reddy, 2001) contents were found to be inhibitory on absorption. Small-sized ligands, such as ascorbic acid, form unstable iron chelates by binding only a few of the 6 coordinating positions of iron, which serve as iron donors to mucin in the duodenal lumen at neutral p H to enhance iron absorption (Benito and Mil le r , 1998). However, dietary constituents such as phytate can form precipitates or stable chelates with iron that interfered with binding between iron and mucin and diminished iron absorption (Gillooly et al, 1983). Removing phytate from bran by dephytinization was shown to yield a bran fraction that had only a small remaining inhibitory effect on iron absorption in human adults (Hallberg et al, 1987). High levels of inorganic calcium and phosphate in bovine milk are also implicated in lowering iron bioavailability (Barton et al, 1983). Due to similarity in physicochemical properties, calcium and iron may compete for common mucosal acceptors preventing movement of iron from the mucosa cell into circulation (Jackson and Lee, 1992). Inorganic phosphate, on the other hand, can oxidize ferrous iron to the ferric state and sequester it as ferric phosphate that is not readily dialyzable (Peters et al, 1971). Dietary proteins can also either enhance or impair iron absorption depending on the food source. While meat proteins enhance iron absorption ( K i m et al, 1995), egg white, bovine milk and soy proteins display an inhibiting effect (Peters et al, 1971; K i m et al, 1995; Cook et al, 1981; Derman et al, 1987). Whole proteins are assumed to be too large to diffuse across the intestinal mucosa and release iron too far along the intestine for efficient absorption. The effect of bovine milk proteins on non-heme iron 27 absorption can be lessened through enzymatic hydrolysis during digestive processes, thereby improving digestive solubility and availability of iron (Hurrell, 1989; Lonnerdal and Glazier, 1989). While protein digestion progresses, more soluble, low molecular weight iron peptides are likely to reach the brush border and enhance iron availability (Slatkavitz and Clydesdale, 1988). Under physiological conditions, it is thought that inorganic forms of iron need to be reduced to the ferrous state for effective absorption. Luminal p H at the site of iron absorption is predicted to be between p H 6.0 and 7.4, whereas calcium phosphate was effective in binding ferrous iron only above p H 5.8 (Peters et al., 1971). The binding of ferrous iron to isolated rat intestinal brush borders was not affected by removal of oxygen, which indicates that the attachment of iron to brush borders does not depend on an energy requiring process (Greenberger et al., 1969). Although protein hydrolysis yields digestion products throughout the upper part of the digestive tract, ferrous iron absorption occurs mainly at the duodenum and proximal jejunum sites of the overall intestine. Therefore, it is useful that iron remains soluble and bound to low molecular weight ligands when it reaches the upper part of the bowel, in order to allow for favorable absorption further down the digestive tract (Ait-oukhatar et al., 2000). The addition of ferrous salts to ocsi- and P-casein causes a rapid uptake of oxygen (Emery, 1992), thus leading to the oxidation of iron and the formation of stable iron-phosphoprotein complexes. In the case of phosphoserine groups in a cluster similar to that of C P P , formation of these complexes is accompanied by the uptake of oxygen in excess to the theoretical requirement for the oxidation of iron to the ferric state (Manson and Cannon, 1978): p C N 1-25 + F e 2 + + 0 2 + 4 H + P C N l -25/Fe 3 + 28 + P C N l -25 /Fe 2 + + 2H2O. This oxygen uptake is absent i f the caseins are first dephosphorylated, thereby indicating the importance of O-phosphoserine groups in the binding of a s i - and P-caseins to iron. Acting as a ligand, phosphopeptide P-casein (1-25) was shown to form a complex with ferrous chloride with a binding capacity of 4 mol Fe 2 + /mole peptide and is responsible for significantly boosting the level of soluble iron at p H 7, from 23 to 81% (Bouhallab et al., 1991). This binding is extraordinarily firm suggesting that CPP-iron interactions involve coordinate bonding in addition to electrostatic bonding. It was proposed that since the affinity of P C N (1-25) to iron is high, compared to calcium iron should be less bio-dispensable and available for absorption (Hurrell et al., 1989). However, binding iron to purified P - C N (1-25), C P P seems to improve iron bioavailability, enhancing erythropoiesis and iron storage in the liver and hastening the cure of iron deficiency in young rats more effectively than ferrous sulfate salt or whole P casein (Ait-oukhatar et al., 1997). P - C N (1-25) can also enhance iron uptake, reduce mucosal storage and improve the net absorption in control and anemic rats using a perfused vascularized duodenal loop model (Peres et al., 1999a). Iron bound to phosphorylated casein is relatively resistant to dissociation; therefore, in this way, iron can be kept in a protected, stable state until it reaches the enterocyte receptor and is released by phosphatases (Pointillart and Gueguen, 1989). Because of the presence of a high concentration of negatively charged phosphate groups, P-CN(l-25)-Fe complex is protected from the action of pepsin and of pancreatic endopeptidases and is hydrolysed to a lesser extent than free P -CN(l-25) during duodenal digestion (Ait-oukhatar et al., 2000). Enzymatic dephosphorylation by phosphatase of 29 the pCN- i ron complex releases free, dialyzable iron, in the ferrous form, at the apical membrane of brush border enterocytes and not in the digestive lumen. The constant association of iron with phosphoserine in contact with the mucous membrane seems to facilitate its transfer at the membrane receptor (Galdi and Valencia, 1988). It was found that fragments of the iron-CPP complex, such as soluble F e / p C N (15-25) with lower molecular weight, reach the brush border more readily than the whole Fe/(3CN (1-25) complex and can better enhance iron absorption (Bouhallab et al., 1999). Binding iron to phosphopeptides keeps it soluble during upper digestive tract transit, preventing against inhibitory interactions with nutrients, such as calcium and zinc, which share similar electrochemical properties and compete for a common transport mechanism at the enterocyte membrane. Although calcium does not seem to be in competition with iron at the rat duodenal brush border membrane vesicle level at a 1:1 ratio, calcium does inhibit iron absorption when present as a 10-fold or 100-fold greater concentration, relative to iron (Wien et al., 1994). The addition of calcium, at an ironxalcium ratio of 1:2, to the perfused intestinal loop of iron-deficient rats significantly lowered iron uptake, but the decrease was lowered by nearly 20% in the |3-CN(l-25) group as compared to the gluconate group (Peres et al., 1997). The presence of zinc also significantly reduced net iron absorption and disappearance from the lumen, lowering uptake by 29-50% at an iromzinc ratio of 1 (Wien et al., 1994). However, when either element was first bound to P -CN(l-25) , no inhibition was observed even when an iromzinc ratio was as high as 1:5 (Peres et al., 1999b). The bound form of trace elements can enhance the digestive absorption by preventing interactions with free ions during membrane transfer, since CPP-Fe and C P P - Z n complexes do not share the same transport 30 mechanisms as the ions. The bound ion complexes, i f absorbed, are likely taken up into the enterocytes by specific pathways such as endocytosis (Peres et al., 1999b). 1.8 Additional functions of CPP 1.8.1 Anticariogenic properties Multi-phosphorylated peptides such as C P P have been proposed to possess anticariogenicity due to an affinity to localize amorphous calcium phosphate in dental plaque, thus maintaining a state of supersaturation with respect to tooth enamel and restricting mineral loss. Although supersaturated with amorphous and crystalline calcium phosphate, stabilization provided by C P P prevents spontaneous precipitation (Reynolds, 1998). C P P bound amorphous calcium phosphate at 0.1% reduces calcium diffusion by 65% at p H 7.0 (Rose, 2000). The amorphous calcium phosphate acts as a buffer for free calcium and phosphate ion species in plaque, thereby preventing tooth enamel demineralization by acid from bacterial plaque in rats (Reynolds et al., 1995) and humans in situ (Reynolds and Riley, 1994) and enhancing re-mineralization in subsurface lesions in human third-molar enamel (Reynolds, 1997). 1.8.2 Immunomodulating properties M i l k is usually a source of host defense for neonates (Cross and G i l l , 2000; G i l l et ah, 2000). Since calcium flux into lymphocytes plays an important part in triggering lymphocyte stimulation, bovine milk derived C P P ' s association with improvement of calcium solubility and utilization has brought many researchers to investigate a possible immunological role. Phosphopeptides, cxsi-casein (59-79) and P-casein (1-25), originated from raw and pasteurized bovine milk, were shown to be responsible for inhibiting concanavalin A -31 (Con A ) induced proliferation of mouse spleen cells and rabbit Peyer's patch cells, while enhancing lipopolysaccharide- (LPS) and phytohemagglutinin- ( P H A ) induced proliferation in these cells (Hata et al, 1998; Kihara and Otani, 2000). In addition to mouse spleen cells and rabbit cells, C P P III, mainly consisting of ctS2-casein (1-32) and 0-casein (1-28), was also shown to have a similar effect on Con A - , L P S - and PHA-induced proliferation of nude mouse spleen cells (Hata et al., 1999). Since L P S , P H A and ConA stimulate different subsets of lymphocytes, namely the B-lymphocytes, helper T-lymphocytes and suppressor/killer T-lymphocytes respectively, it was shown that C P P suppresses killer T-cells, while also playing a role in boosting B-lympocytes and helper T-lymphocytes (Hata et al, 1999). Mitogenic activity of CPP-III on mouse spleen cells was unaffected by prior digestion with pepsin, trypsin/chymotrypsin or pancreatin, but was greatly severed when C P P was previously treated with potato acid phosphatase (Hata et al, 1999). Therefore, it can be concluded that mitogenic activity in cell culture is attributable to the phosphoserine residues. When cultured with mouse spleen cells, C P P a si-casein (59-79) also significantly enhanced the immunoglobulin levels of IgG, I g M and IgA (Hata et al, 1999). Oral ingestion of CPP-III has been shown to be beneficial for the enhancement of mucosal immunity in mice. Mice fed CPP-III supplemented diets at 0.1 to 1.0% had a significantly greater intestinal specific IgA response towards antigens such as ovalbumin and P-lactoglobulin (Otani et al, 2000). Although C P P seems to be quite a promising and beneficial immunostimulant as a dietary supplement, it must be utilized with caution since the multi-phosphorylated region 32 of phosphopeptides is also identified to be the epitope for anti-casein antibodies binding (Perich et al, 1999). The antigenicity of C P P was established by Park and Al len (2000) in a rat model. Tryptic casein phosphopeptides from a- and P-caseins were shown to induce a significantly lower IgG specific response (to whole a- or P-casein) than intact proteins or whole tryptic hydrolysates. Enzymatic hydrolysis of caseins served only to lower but could not remove antigenicity entirely. 1.9 Milk Protein Antioxidants Mineral supplementation to milk as a boost to the nutritional quality may also put the milk system at risk towards oxidative changes, i f bound to the milk fat fraction. For example, iron (II) can undergo cyclic autoxidation and form ferric hydroxide-protein complexes that may contribute to further instability to supplemented milk during prolonged storage, thereby severely diminishing the nutritional quality for neonates (Gutteridge and Hal l iwel l , 1994). In order to counteract this oxidative vulnerability, the role of antioxidants is fundamental in the milk system (Lindmark-Mansson, 2000). Hal l iwel l (1995) defined antioxidant simply as "any substance that, when present at low concentrations compared with those of an oxidizable substrate, wi l l significantly delay or prevent oxidation of that substance". Oxidizable substrate refers to any molecule found in foods and in l iving tissues, such as lipids, proteins, carbohydrates or D N A , that are vulnerable to attack by Reactive Oxygen Species (ROS). These include the superoxide anion (O2'), hydroxyl radical ( O H ) , and oxygen-centered radicals of organic compounds (peroxyl R O O ' and alkoxyl RO') together with non-radical reactive compounds such as hydrogen peroxide (H2O2) and singlet oxygen (^2) (Morrissey and O'Br ien , 1998). 33 In bovine milk, major protein fractions including whey and casein have been implicated in displaying various antioxidative characteristics. A non-heme iron-binding glycoprotein in the whey fraction of bovine milk, lactoferrin, has an affinity for iron and inhibited iron-catalyzed oxidation in iron-supplemented infant formulas (Satue-Gracia et al., 2000) and liposomal phospholipid systems (Wakabayashi et al., 1999). A high molecular weight fraction of whey was also shown to act as a scavenger for peroxyl radicals in salmon oil-in-water emulsion systems (Tong et al., 2000). Similarly, phosphate-bound phosphoserine residues in casein have the ability to scavenge superoxide, hydroxyl and l,l-diphenyl-2-picrylhydrazyl (DPPH) radicals (Suetsunna et al., 2000) and to lower oxidation in iron-induced peroxidation of arachidonic acid in liposomes (Cervato et al., 1999). Antioxidants are classified into three different categories: Primary antioxidants donate electrons and terminate free radical chain reactions through converting l ipid radicals to more stable products; secondary antioxidants lower the rate of chain initiation by a variety of mechanisms: acting as an reducing agents, chelating agents for metallic catalysts such as copper and iron or as singlet oxygen quenchers; enzymatic antioxidants remove highly oxidative species such as hydrogen peroxide or superoxide (Gordon, 1990). The most common synthetic chain-breaking antioxidants used in foods are butylated hydroxyanisole ( B H A ) , butylated hydroxytoluene (BHT) , propyl gallate (PG) and nordihydroguaiaretic acid ( N D G A ) . The relatively low cost, superb efficiency and stability allow these synthetic antioxidants to lower oxidation. However, repeated concerns about the safety (i.e. possible carcinogenic effects) of these antioxidants have led to a growing interest in 'natural" antioxidants. Ideally, antioxidants should be found 34 "naturally" in l iving organisms, plants and animals inclusive, for example, tocopherol in vegetable oils, ascorbic acid in fruits (Gutteridge and Hal l iwel l , 1994) or even bioactive peptides derived from milk (Meisel, 1998). 1.10 Lipid Auto-oxidation and Fenton Reaction The most important stressors in aerobic organisms and food systems are the reduced derivatives of oxygen, free radicals such as the reactive oxygen species, which exist naturally either as a part of normal physiological and metabolic processes or as a product of food processing and manufacturing. It has been established that oxidative deterioration of lipids are caused by auto-oxidation reactions, which are essentially free radical chain reactions proceeding via defined initiation, propagation and termination steps. In the initiation step, hydrogen atoms are abstracted from the methylene (-CH.2-) group between double bonds in polyunsaturated fatty acids generating free radicals that are stabilized by resonance (Richardson and Korycha-Dahl, 1983). The following reactions summarized the steps auto-oxidation: Initiation: L H —> L ' [1] L ' + 0 2 ~ > LOO* [2] Propagation: L O O ' + L ' H —> L O O H + L [3] Termination: L O O ' + L ' ' —> L O O L ' [4] L O O ' + L ' O O ' —> L O O L ' + 0 2 [5] L ' + L ' ' — > L L ' [6] 35 In initiation (e.g. Equations 1 and 2), the fatty acid (LH) forms a free radical (L'), which reacts with oxygen to from the peroxy radical (LOO') . The oxygenation of L is very rapid and is therefore the primary product of oxidation. Oxidation can also be initiated by one electron species such as transition metals with valence states one unit apart (e.g. F e 3 + and Fe 2 + ) (Larson, 1997). These radicals react with other unsaturated fatty acids by abstracting more hydrogen atoms in the propagation step (Equation 3). Reactions during the propagation step become self-catalytic and accelerate until the free radicals become stabilized during termination (Equations 4, 5 and 6). Peroxides, formed as a result of lipid auto-oxidation, are then subsequently transformed into a variety of oxygenated substances, such as alcohols, hydroperoxides, ketones, epoxides and others reactive products, which undergo further reactions that yield a complex and potentially toxic mixture of polymers (Shahidi, 1997). In the presence of metals such as iron (II), l ip id autoxidation can be induced through the Fenton reaction, which involves the oxidation of iron [Equation 7], generation of hydroxyl [Equation 7] and superoxide radicals [Equations 8 and 9]. F e 2 + + H 2 0 2 —> F e 3 + + ' O H + OH" [7] F e 3 + + H 2 0 2 —> F e 2 + + 0 2 " + 2 H + [8] ' O H + H 2 0 2 —> H 2 0 + H + + 0 2 " [9] Reduced iron catalyzes the formation of hydroxyl radicals and of other reactive oxygen species via the Fenton reaction, thereby causing free radical-mediated l ipid peroxidation. H 2 0 2 is non-toxic and relatively stable as a molecule by itself, but can become a precursor of the highly reactive ' H O , which oxidizes all substances close to it. 36 Since it is nearly impossible to completely abolish the presence of initiating species such as preformed free radicals and other pro-oxidative factors such as metal ions (Richardson and Korycka-Dahl, 1983), antioxidant agents should target at stabilizing reactive species generated in the propagation phase to effectively inhibit or slow l ipid oxidation in real food systems and in vivo. In general, iron supplementation in food sources such as bovine milk can improve the bioavailability of iron for humans, but may also increase the exposure of reactive oxygen species generated by the iron-induced Fenton reaction to both the food system and the consumer. Since iron excretion from the body is limited and largely unregulated, iron is potentially more toxic than other trace elements due to the lack of effective means to protect human cells against iron overload and the role of iron in generation of free radicals. It was found that intestinal epithelial cells play a key role in regulating body iron absorption and intracellular iron levels (Arredondo et al., 1997). A n excess of free dietary iron is associated with oxidative damage in D N A and lipid peroxidation at biomembranes (Wessling-Resnick, 2000), which lead to erosion of microvil l i at the gastrointestinal (Gl) tract absorptive surface and eventual mucosal cell apoptosis and necrosis (Srigiridhar et al, 2001). In addition to damage of the G l tract, prolonged iron overload, and the associated accumulation of oxidation products in tissues, may lead to conditions such as liver failure, diabetes, heart malfunction and joint inflammation (Halliwell and Gutteridge, 1984). 1.11 Protein Glycosylation Since proteins and bioactive peptides derived from bovine milk have been shown to display a variety of functionalities, it is of interest so determine i f these functionalities 37 can be improved by different modification techniques. Various functional properties of proteins have been shown to be improved by conjugation with polysaccharides or oligosacchardies. The conjugation reaction involves covalently binding e-amino groups of the protein to the reducing-end carbonyl groups of the sugar, under controlled heating and humidity conditions, similar to that in a Maillard reaction (Figure 2). These protein-saccharide conjugates are also known as neoglycoproteins. Polysaccharide Protein Glyocwylprotein (protein- r*d y saccharide conjugate) Figure 2. Protein-polysaccharide conjugate product using Maillard Reaction. (Nakamura et al., 1998) To date, research on the modification of bovine milk proteins by glycosylation is scarce, but many egg proteins have been shown to display enhanced functionalities, following glycosylation with saccharides. Covalently binding egg white protein, ovalbumin, to dextran, a large neutral polysaccharide, was shown to aid in emulsification unto l ipid surfaces, hence more effectively scavenging free radicals there and improving antioxidant activities (Kato et al., 1988; Kato et al., 1990; Nakamura et al., 1992). Enhanced emulsifying properties of lysozyme via glycosylation with dextran are suggested to contribute to greater antimicrobial activity targeted towards the lipopolysaccharide outer membrane in Gram-negative bacteria (Nakamura et al, 1991). In addition to dextran, antioxidant activity of ovalbumin, antimicrobial activity of lysozyme (Nakamura et al., 1994; Nakamura and Kato, 2000) and emulsifying properties 38 of phosvitin (Sattar Khan et al., 1999; Nakamura et al., 1998) were also significantly enhanced through conjugation of the proteins with galactomannan, a polysaccharide derived from guar gum. The functionalities of these protein-saccharide conjugates may be affected by the length and number of saccharide chains attached. Emulsifying properties and heat stability of lysozyme-galactomannan conjugates improved with the length and number of polysaccharide chains, whereas these functionalities were not affected by conjugation with an oligosaccharide, xyloglucan (Shu et al., 1996). These researchers then concluded that conjugation with long, branched chain polysaccharides such as dextran and galactomannan efficiently improved functionality of proteins but this improvement was not as apparent when proteins are conjugated with shorter chain or straight chain oligosaccharides. However, these findings were refuted by a five fold higher emulsifying activity in lysozyme-oligosaccharide (i.e. xyloglucan) conjugates when compared to the native proteins (Nakamura et al., 2000). Since bovine milk P-casein was found to have higher surface activity and superior adhesion to oil-in-water emulsions than milk whey protein P-lactoglobulin and egg yolk phosvitin (Dickinson et al., 1991), addition of saccharides to peptides derived from casein may serve to amplify the hydrophilicity of the complex, enhancing surface activity and possibly improving metal chelating and free radical scavenging activities in aqueous and emulsion models previously observed in these peptides. Since these peptide-saccharide conjugates are not likely to be toxic for ingestion (Nakamura and Kato, 2000), they have a great potential to be bioactive in vitro and in vivo. 39 C H A P T E R II: O U T L L I N E O F R E S E A R C H ACTIVITIES 2.1 General Introduction As a potent bioactive peptide with metal binding characteristics, bovine milk casein-derived caseinophosphopeptides (CPP) may serve to remove pro-oxidative metal catalysts such as iron from an environment of oxidizable lipids (Hegenauder et al., 1979), from dietary sources or at biomembrane lipid bilayer systems of potential target organs. Since the absorption of iron is mainly regulated at the gastrointestinal tract, it is important to search for potential iron chelators that have a selective affinity for iron, exhibit a low toxicity in cells upon oral intake (Faa and Crisponi, 1989) and, most importantly, remain bioactive under gastrointestinal alkaline conditions (Benito and Miller, 1998). Since bovine milk P-casein was found to have higher surface activity and superior adhesion to oil-in-water emulsions than milk whey protein p-lactoglobulin and egg yolk phosvitin (Dickinson et al., 1991), conjugation of saccharides to peptides derived from casein may serve to amplify the hydrophilicity of the complex, enhancing surface activity and possibly improving antioxidant activities in aqueous and emulsion models. Since these peptide-saccharide conjugates are not likely to be toxic for ingestion (Nakamura and Kato, 2000), they have a great potential to be bioactive in vitro and in vivo. The following in vitro studies were designed to examine the potential of bovine milk caseinophosphopeptides (CPP) to serve as an antioxidant, through metal chelating and/or free radical sequestering, in aqueous, emulsion and cell culture models. In addition, the effect of glycosylation on the CPP antioxidant activity was investigated. 40 2.2 N u l l Hypothesis (H 0 ) 1. CPP has no affinity to sequester divalent metals such as iron and calcium. 2. CPP does not behave as an antioxidant by sequestering metal catalysts and inhibiting the Fenton Reaction. 3. CPP does not exhibit both primary and secondary antioxidant activity. 4. Conjugation between CPP and saccharides will not alter antioxidant activity. 5. CPP is non-toxic to enterocytes. 6. CPP has no role in inhibiting oxidation reactions at the site of the human intestine by sequestering free metals in the intestinal lumen. 2.3 Objectives 1. To determine the affinity of CPP for sequestering iron (II), (III) and calcium. 2. To determine if CPP has the affinity to inhibit Fenton reaction-induced oxidation in various in vitro model systems. 3. To determine if CPP protects against oxidation reactions induced by stable free radicals or a free radical initiator. 4. To determine if the affinity of CPP in the above models is altered by conjugation with galactomannan and xyloglucan. 5. To determine if CPP can protect human intestinal carcinoma cells (Caco-2) and embryonic cells (Int 407) from bivalent metal-induced cytotoxicity. 2.4 Exper imental Approach 1. To compare absorbance scans (200-600nm) for CPP, divalent metals and CPP-metal complex for the purpose of estimating CPP-mineral interaction and possible sequestering activity. 41 2. To perform site-specific binding and non site-specific binding assays with deoxyribose and liposome assays, using iron as an initiator for oxidation. 3. To use free radicals as an initiator for oxidation and perform stable free radical D P P H and A B T S assays and liposome assay (with A A P H ) to assess primary antioxidant activity. 4. To assess the viability of human intestinal carcinoma cells (Caco-2) and embryonic cells (Int-407) following exposure to ferrous iron-induced toxicity by the Mitochondrial Tetrazollium (MTT) Test. 2.5 Comments on the methodologies for antioxidant analysis In order to detect an interaction between C P P and metal ions, C P P wi l l be incubated with the following metal ions (i.e. iron II, iron III and calcium) in a 37°C water bath for one hour prior to measuring for absorbance. Absorbance scanning between 200-600 nm wi l l provide a quick and convenient method for identifying the interaction and possible sequestering between ligand and ion (Hu and Kitts, 2000). If the scan for the CPP-metal complex shows peaks not present for C P P and metal ions individually, then initial evidence for binding is obtained. The most common methods to monitor the affinity of a potential antioxidant usually involves inducing l ipid auto-oxidation in bulk oi l or emulsion systems. Of all the reactive oxygen species, the hydroxyl radical is the most potent oxidant in biomolecules, making it extremely effective in starting l ipid peroxidations. In the presence of divalent metal ions (Fe 2 + ) , fatty acids can be oxidized to produce peroxyl radicals ( R O O ) , which then undergo a complex series of reactions to yield alkoxyl radicals (RO') and other 42 degradation products such as malondialdehyde- ( M D A ) and conjugated diene-containing substances (Gutteridge and Hall iwel l , 1990). The rate constants for hydroxyl radical-induced reactions can be determined in a simple test tube assay, the deoxyribose assay. Rate constants for radical-induced oxidation reactions had been measured previously by highly expensive methods, such as pulse radiolysis. However, this inexpensive deoxyribose assay gives accurate results despite the simplicity (Halliwell et al., 1987). The sugar 2-deoxy-D-ribose is degraded (Equation 13) on exposure to hydroxyl radicals generated by the Fenton reaction (Equations 10, 11 and 12). If the resulting complex mixture of products is heated under acid conditions, a secondary product of oxidation, malondialdehyde ( M D A ) , is formed and this may be detected by the reaction with thiobarbituric acid ( T B A ) to form a pink chromogen (Equation 14). The deoxyribose assay can therefore be a useful method for measuring the formation of O H radical in biochemical systems. F e 2 + - E D T A + 0 2 <~> F e 3 + - E D T A +0" 2 [10] 20" 2 + 2 H + H 2 0 2 + 0 2 [11] F e 2 + — E D T A + H 2 0 2 -» OH" + ' O H + F e 3 + — E D T A [ 12] ' O H + deoxyribose -> fragments -> heat & acid -> M D A [13] 2 T B A + M D A -> chromogen [14] There are two different ways of determining antioxidant activity using this assay. If deoxyribose is incubated with iron in the presence of E D T A , the formation of hydroxyl radicals ( O H ) is favored, accelerating the oxidation of the sugar. This is regarded as the non-site specific binding assay, since C P P , when added, w i l l interact with hydroxyl 43 radicals generated rather than directly with the deoxyribose. On the other hand, when E D T A is replaced with an equal volume of phosphate buffer, the assay can detect site specific binding when C P P and deoxyribose compete for the metal iron. Reduction in deoxyribose degradation is indicated by a lowering in absorbance values at 532 nm, for both types of assay. The potential antioxidant effect of C P P wi l l also be determined in emulsion models, namely phospholipid liposomes. Liposomes were first described by Bangham (1965) while studying the nature of cell membranes. It was found that liposomes formed spontaneously when phospholipids were dispersed into water. Liposomes are simply spherical vesicles consisting of one or more phospholipid bilayers surrounding an aqueous cavity. Initial studies with liposomes focused on the cell membrane; however, the majority of studies now use liposomes for drug delivery research (Rongen et al., 1997). The liposomes used in the following studies were formed by dissolving soybean lecithin in a saline phosphate buffer and used as an emulsion system to test the extent of oxidation induced by metals or free radicals. The extent of oxidation was monitored by the formation of conjugated diene products arising from lipid hydroperoxides, l ipid alcohols and fatty acids, which contain no oxygen other than the carboxyl group (Smith and Anderson, 1987). Measurement of conjugated diene at an absorbance of 234 nm is a fast, accurate and convenient method for determining the relative amount of l ipid peroxidation. Oxidation can be initiated by the addition of ferrous iron (Fell) or peroxyl radicals generated by the thermolysis of 2,2-azobis(2-amidino-propane) dihydrochloride ( A A P H ) . The reaction system should be maintained for 100 minutes at 37°C, with 44 measurements taken every 4 minutes. The antioxidant potency is determined from the reduction in peroxidation propagation rate, as compared to the control. A second method to detect the free radical scavenging potential of C P P is through the reaction with a stable free radical cc,oc-diphenyl-P-picrylhydrazyl (DPPH). Because of the odd electron in D P P H , it wi l l show a strong absorption band at 517 nm in ethanol. The solution thus appears deep violet (Blois, 1958). If the electron is paired off, the absorption diminishes and the resultant decolorization correlates well with the number of electrons taken up. The test agent donating the electron or hydrogen atom would therefore be an active antioxidant. Since the solvent of the D P P H is ethanol, a medium where C P P is not very soluble, an alternate stable radical 2,2-azinobis-(3-ethylbenzothiazoline-6-sulfonic acid) A B T S - + , which is more hydrophilic can also be used. The pre-formed radical monocation A B T S - + is generated by oxidation with ammonium persulfate, resulting in the production of a blue/green chromophore with an absorption maxima of 734 nm (Pellegrini et al., 1999). The addition of antioxidants to the pre-formed radical cation would reduces A B T S ' + to A B T S on a time-scale, which is dependent on the antioxidant activity and concentration of the test sample. The extent of decoloration, when expressed as % inhibition of the A B T S ' + , can be monitored by the change in absorbance at 734 nm. The human colon adenocarcinoma cell (Caco 2) culture model has been exploited recently as a model for measuring bioavailabilities of iron from foods (Garcia et al., 1996), calcium (Ekmekcioglu, 1999) and zinc (Hansen et al., 1996), due to an affinity to differentiate spontaneously into biopolar enterocytes that exhibit many of the characteristics of normal epithelial cells such as microvil l i junctions and the excretion of 45 brush border associated enzymes (Halleux and Schneider, 1994). However, since Caco-2 cells are carcinoma cells, metal-induced cytotoxicity in the Caco-2 cells in addition to non-carcinomic human embryonic intestinal cells (Int-407) wi l l be evaluated for a comparison. Cytotoxicity can be measured using Mitochondrial Tetrazollium (MTT) Test (Mosmann, 1983). Viable cells with a functional mitochondrial enzyme cleave the mitochondrial tetrazollium salt into a blue colored formazan. The amount of formazan produced, which correlates to the number of viable cells, is monitored spectrophotometrically (A570). 46 C H A P T E R III: ANTIOXIDANT E F F E C T S O F C O M M E R C I A L L Y A V A I L A B L E CPP 3.1 Materials and Methods Ferrous chloride, calcium chloride, L-ascorbic acid, 2 deoxy-D-ribose, 2 thiobarbituric acid ( T B A ) , l,l-diphenyl-2-picrylhydrazyl (DPPH), L-a-lecithin (from soybean), 2,2'-azinobis (3-ethylbenzothiazoline-6-sulfonic acid) (diammonium salt, A B T S ) , chelex-100 ion-exchange resin, and casein were obtained from Sigma Chemical Co. (St. Louis, M O ) . Ferrous sulfate, E D T A , hydrogen peroxide (30%) and trichloroacetic acid ( T C A ) were from Fisher Scientific (Fairlawn, NJ). Methanol, acetic acid and hydrochloric acid were from Fisher Scientific (Napean, ON) . Ferric chloride was purchased from Mallinckrodt (Paris, Kentucky). Sodium hydroxide, potassium dihydrogen orthophosphate and sodium chloride were obtained from B D H Inc. (Toronto, ON) . 2,2'-azobis (2-amidinopropane) dihydrochloride ( A A P H ) was from Wako Chemicals U S A Inc. (Richmond, V A ) . Sodium dodecyl sulfate (SDS), ammonium persulfate, glycine, coomassie blue G-250, broad range molecular weight standards, tris base, (3-mercaptoethanol, glycerol and bromophenol blue were from Bio-Rad Laboratories (Richmond, C A ) . Phosphate buffer was made with distilled deionized water and was eluted through a Chelex-100 column to eliminate the occurrence of transition metals. 3.1.1 CPP Formulations Caseinophosphopeptides (CPPms) were obtained from Mei j i Seika Kaisha Ltd. (Kawasaki, Japan). CPPms-I and CPPms-III were commercially produced by treating a whole casein solution [15% (w/v), p H 7.0] with trypsin (Novo Industri Japan) at 60°C for 2 hours to give the hydrolysates (Naito et al., 1990). Enzyme hydrolysis for production 47 of CPPms-I and CPPms-III required 0.002% (w/w) and 0.01% of the substrate respectively. CPPms-I was obtained from spray-drying the whole hydrolysate. CPPms-III enriched with phosphopeptides was produced by filtering the insoluble mass from the hydrolysate at p H 4.5 before precipitating the phosphopeptides by adding calcium chloride [1.1% (w/v)] and ethanol [50% (v/v)]. 3.1.2 CPP Characterization Studies 3.1.2.1 Molecular Weight Characterization Casein, CPPms-I and CPPms-III were analyzed by S D S - P A G E according to the method of Laemmli (1970) using 15%T (i.e. weight percentage of total monomer including crosslinker) polyacrylamide separating gel and 4% stacking gel containing 0.1% SDS on the Mini-Protean I M i n i - C e l l slab gel electrophoresis unit ( B I O - R A D Laboratories, Richmond, C A ) . Protein samples (40jug) were heated at 100°C for 5 minutes in 0 . 5 M Tris-glycine buffer, p H 6.8 containing 10% SDS. Electrophoresis was carried out at a constant voltage of 100 volts for 1.5 hours with a Tris-glycine running buffer. Gels were placed in 0.1% Coomassie Blue G-250 staining solution (40% methanol and 10% acetic acid) for 30 minutes and destained in 40% methanol and 10% acetic acid for 3 x 15 minute destain washes. 3.1.2.2 Iron and Calcium Chelating Capacity A spectrophotometric method was adopted from H u and Kitts (2000) to estimate the metal ion chelating capacities of CPPms. CPPms-I and CPPms-III ( lmg/ml) were incubated with 100 LiM F e C l 2 , F e C l 3 or C a C l 2 at 37°C for 2 hours. The differential spectra of CPPms, metal chlorides, and CPPms-metal chloride complexes were recorded against 10 m M phosphate buffer (pH 7.4) over a wavelength range of 200-600 nm. 48 3.1.3 Determination of Antioxidant Activity 3.1.3.1 Hydroxyl radical scavenging assay in a Deoxyribose model A Fenton reaction model containing 0.1 m M of F e 3 + as the catalytic metal was used. The substrate 2-deoxyribose (3.6 m M ) was mixed together with CPPms (0.05, 0.10, 0.50, 1.00 mg/ml) and 0.1 m M E D T A , 0.1 m M ferric chloride, 0.1 m M ascorbic acid and I m M H2O2. The reaction mixture was incubated at 37°C for 1 hour. Following incubation, 1 ml of 10% (v/v) trichloroacetic acid ( T C A ) and 1% 2-thiobarbituric acid ( T B A ) were added and the mixture was boiled for 15 minutes at 95°C. Absorbance at 532 nm was recorded after cooling. The extent of deoxyribose degradation by hydroxyl radicals generated by the Fenton reaction was calculated using the equation: % Inhibition = A b s @ 5 3 2 n m m n t m i - Abs@532nnu a m p ie x 100 Abs@532nm c o n troi 3.1.3.2 Peroxyl Radical and Metal scavenging assays in a Liposome model Liposomes were made by sonicating soybean lecithin (cc-phosphatidylcholine) in an ice-water bath for 20 minutes in 10 m M phosphate buffer (pH 7.4) according to H u and Kitts (2000). Peroxyl radical induced liposomal peroxidation was conducted at a constant temperature of 37°C using an A T I Unicam (UV2) U V / V i s spectrophotometer. The reaction was initiated by the addition of 0.2 m M A A P H to a mixture of 0.1 mg/ml of liposome in 10 m M phosphate buffer (pH 7.4) and C P P . Generation of conjugated diene hydroperoxide was monitored by taking absorbance readings at 234 nm every 4 minutes for 100 minutes. 3.1.3.3 Free Radical Scavenging Activity The affinity of CPPms to scavenge a stable l,l-diphenyl-2-picrylhydrazyl (DPPH) was measured according to Blois (1958). CPPms (0.05, 0.10, 0.50, 1.00 mg/ml) 49 was mixed with 0.1 m M D P P H radical in ethanol solution prior to an incubation period of 20 minutes at room temperature. Discoloration was monitored by measuring the absorbance at 519 nm following incubation. The (2,2'-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) A B T S radical cation model was also used to evaluate the free radical scavenging effect of CPPms (Pellegrini et al, 1999). Ethanol was replaced by distilled deionized water in the radical preparation. Discoloration is determined by comparing the absorbance at 734 nm of the treatment groups with the control after an 8-minute incubation at room temperature. The percentage inhibition in the D P P H and A B T S assays was calculated according to the following equation: % Inhibition = A b s © 5 1 9 / 7 3 4 n r r w m i - Abs@519/734nm , a m pi P xl00 Abs@519/734nm c o n t r o l 3.1.4 Statistical Analysis A l l data (with the exception of the liposome assays) were collected in triplicates and analysed by one way A N O V A (a < 0.05), followed by a multiple range Tukey post test analysis using the GraphPad Prism Analysis software (GraphPad Software Inc., San Diego C A ) to identify significant difference among treatment means (P < 0.01). Data were obtained from 3 different individual experiments for the liposome assays and submitted to linear regression using GraphPad software. Rates of propagation (i.e. slope of regression equation) were then analyzed by one way A N O V A (a < 0.05) and Tukey test (P < 0.01). 50 3.2 Results 3.2.1 Molecular Characterization of CPP Caseinophosphopeptides (CPPms-I and -III), purchased from Mei j i Seika Kaisha Ltd. , were analyzed using SDS P A G E electrophoresis to estimate the apparent molecular weight of peptides present (Figure 3). Individual protein bands for the whole casein sample had apparent molecular weights of approximately 22-23 kDa, 25 kDa, 23-34 kDa and 19 kDa, respectively. These sizes are consistent with the reported molecular weights of a s i , aS2, P and K respectively (Swaisgood, 1982). As for the CPPms-I and CPPms-III samples, a number of peptides having a molecular weight of lower than 6 kDa was observed. This electrophoresis method used did not however provide a precise description of the exact molecular weights of small peptides. Other workers have reported C P P preparations consisting of 40% a s r C N ( 4 3 - 7 9 ) and 36% P -CN(l -25) with 4.6 kDa and 3.1 kDa respectively (Naito et al, 1990; Reynolds et al, 1994; Ono et al, 1998; Gagnaire et al, 1996). 51 Figure 3. SDS P A G E G e l (15%T) using the Laemml i buffer system (0.375 M Tr i s , p H 8.8). Lane 1: 40 ug Casein, Lane 2: 40 ug CPPms-III, Lane 3: 40 ug CPPms -I , Lane 4: B r o a d Range Molecular Weight M a r k e r . 52 3.2.2 Qualitative M e t a l Ion Sequestering Act iv i ty of C P P m s Absorbance scanning between 200 and 600 nm was performed to detect interactions between CPPms and metal ions of interest. Absorbance scans were performed (Figures 4 to 9) for Fe (II), Fe (III), Ca (II), CPPms-I, CPPms-III and the different CPPms-metal mixtures following one hour incubation at 37°C. The differential spectra for each metal and CPPms revealed characteristic absorbance peaks, whereas peaks for CPPms-metal mixtures resembled neither the metal ion nor the peptide, indicating that an interaction between these two substances had occurred. Although the strength of binding cannot be quantified from the results, the spectra provided qualitative evidence for an interaction between CPPms and F e 2 + or F e 3 + , while also distinguishing the binding characteristics between CPPms-I and III. 53 2 6 0 . 0 1 0 0 . 0 ( N M ^ D I U . > 6 0 0 . 0 Wavelength (nm) Figure 4. Differential spectra of CPPms-I with ferrous iron: Spectrum 1 represents the spectrum of CPPms-I against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-I and Fe 2 + against phosphate buffer; spectrum 3 represents the spectrum of Fe 2 + against phosphate buffer. 54 + 2 . 00A / * " " * \ 3 \ / 0.50 0 {Av-DIU . > , V Absorbance + 0 . 00A 1 1 1 ' — 200.0 100.0(NM/DIU.> 600.0 Wavelength (nm) Figure 5. Differential spectra of CPPms-I with ferric iron: Spectrum 1 represents the spectrum of CPPms-I against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-I and Fe + against phosphate buffer; spectrum 3 represents the spectrum of Fe 3 + against phosphate buffer. 55 2 0 0 . 0 1 0 0 . 0 < N M / Q I U . > 6 0 0 . 0 Wavelength (nm) Figure 6. Differential spectra of CPPms-I with calcium: Spectrum 1 represents the spectrum of CPPms-I against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-I and C a 2 + against phosphate buffer; spectrum 3 represents the spectrum of C a 2 + against phosphate buffer. 56 Figure 7. Differential spectra of CPPms-III with ferrous iron: Spectrum 1 represents the spectrum of CPPms-III against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-III and Fe 2 + against phosphate buffer; spectrum 3 represents the spectrum of F e 2 + against phosphate buffer. 57 Figure 8. Differential spectra of CPPms-III with ferric iron: Spectrum 1 represents the spectrum of CPPms-III against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-III and F e 3 + against phosphate buffer; spectrum 3 represents the spectrum of Fe 3 + against phosphate buffer. 58 + 2 . 0 6 A 0 . 5 0 0 < A s D I U . > Absorbance + 0 . 0 0 A 100 . 0X N M D I U Wavelength (nm) NM 6 0 0 . 0 Figure 9. Differential spectra of CPPms-III with calcium: Spectrum 1 represents the spectrum of CPPms-III against lOmM phosphate buffer (pH 7.4); spectrum 2 represents the spectrum of a mixture of CPPms-III and C a 2 + against phosphate buffer; spectrum 3 represents the spectrum of C a 2 + against phosphate buffer. 59 3.2.3 Site-Specific and non site-specific scavenging activity of CPPms The affinities of CPPms to sequester iron and quench hydroxyl radicals were evaluated using two different assays in a deoxyribose model. In the site-specific assay (e.g. F e 2 + + H2O2), the affinity of CPPms to chelate iron and minimize the direct interaction between the pro-oxidative iron and deoxyribose is tested. With the addition of E D T A in the non site-specific assay (e.g. F e 2 + - E D T A + H 2 0 2 ) , direct F e 2 + binding to deoxyribose is restricted which favors the formation of O H in the aqueous medium. In the latter assay, the affinity of CPPms to quench hydroxyl radicals was tested. CPPms was found to be highly effective in suppressing radical induced damage of deoxyribose in both assays (Table 4; Figures 10a and 10b). This result suggests that not only does C P P possess secondary antioxidant characteristics, as a metal chelator, but CPPms can also act as a primary antioxidant via direct interference with free radicals. Both samples of CPPms exhibited a concentration dependent increase in activity. CPPms-I inhibited up to 50% of the deoxyribose degradation, at 1.0 mg/ml, in both assays, whereas CPPms-III, at l.Omg/ml, lowered deoxyribose degradation by 50% in site-specific assay, but only 25% in the non site-specific assay. In the non site-specific assay, CPPms-I was significantly more effective in inhibiting the deoxyribose oxidation than CPPms-III (p<0.01) at the 2 highest concentrations (i.e. 0.5 and 1.0 mg/ml). 60 Table 4. Effect of CPPms-I and -III on Percent Inhibition of Deoxyribose Degradation 1. Concentration of C P P (mg/ml) Assay CPPms 0.05 0.10 0.50 1.00 I 33.6 ± 1.7ax 34.2 ± 3 .3 a x 42.2 ± 1.5 a b x 46.2 ± 1.8b x Site-Specific Binding 2 III 27.1 ± 0 . 9 a x 32.8 ± 0 .8 a b x 40.5 ± 1.4 b c x 45.5 ± 0 . 1 c x I -0.1 ± 0 . 1 a x - 0 . 2 ± 0 . 1 a x 18.4 ± 3 .1 b x 43.7 ± 1.7CX Non Site-Specific Binding 3 III - 0 . 5 ± 0 . 1 a x - 0 . 5 ± 0 . 1 a x 3 . 6 ± l . l a y 25.5 ± 0.4 b y 1. % Inhibition = Abs 5 3 2 m r w r n i - Abs 532nm s n m pip. xlOO; Abs 532nm c o ntroi Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (e.g. a, b, c and d) are significantly differently (p<0.01); Means within the same column for each assay that do not share a common superscript letter (e.g. x and y) are significantly differently (p<0.01) 2. Ascorbic acid mediated Fenton Reaction in the absence of E D T A 3. Ascorbic acid mediated Fenton Reaction in the presence of E D T A 61 60 n 50 A 0.05 0.10 0.50 1.00 Concentration of CPPms (mg/ml) Figure 10a. Effect of CPPms-I ( • ) and CPPms-III ( • ) on Percent Inhibition of Deoxyribose Degradation (Site-Specific Assay)1. 1. % Inhibition = Abs 5 3 2 n m m n t m i - Abs 532nnu i m pie. xlOO; Abs 532nmCOntroi Results are expressed as Mean ± SD, n = 3; Reaction initiated by ascorbic acid mediated Fenton Reaction in the absence of E D T A . 62 Figure 10b. Effect of CPPms-I ( • ) and CPPms-III ( • ) on Percent Inhibition of Deoxyribose Degradation (Non Site-Specific Assay)1. 1. % Inhibition = Abs 532nmfn„tmi - Abs 532nm B a j B E i £ xlOO; Abs 532nm c o n t r o i Results are expressed as Mean ± SD, n = 3; Reaction initiated by ascorbic acid mediated Fenton Reaction in the presence of E D T A . 63 3.2.4 Effect of CPPms on the formation of Conjugated Diene The effect of CPPms in protecting lecithin liposomes (i.e. an emulsion system) from oxidation, induced by pro-oxidative free radicals or bivalent metals, was determined by monitoring the formation of conjugated diene at an absorbance of 234 nm. The rates of propagation, expressed as percent of control, were calculated by applying linear regression to the kinetic plots of absorbance (A234nm), against time of incubation (minutes) and comparing the slopes obtained for each treatment (Appendix; Figures A l and A2) . Inhibition in the rate of propagation was most significant (p<0.01) at the highest concentration of CPPms used (i.e. 0.50 mg/ml). CPPms-I found to be two fold more effective than CPPms-III (p<0.01) in quenching the free radical A A P H at the liposomal interface, lowering oxidation to 23.0 % control as compared to 43.3 % control for C P P m s - m . The addition of ferrous ions initiated liposome oxidation possibly through an intricate series of free radical chain reactions. Since plots of absorbance against incubation time showed a more rapid progression of oxidation, an alternate method was used to analyze the kinetics of propagation. Percent inhibition was calculated at each time point of the plot to achieve a plateau (Appendix; Figures A 3 and A4) and the y-intercept was used for comparison (Table 5). In this assay, CPPms-I and CPPms-III were not significantly different in activity (p>0.05) at the highest concentration (i.e. 0.50 mg/ml) and lowest concentration (i.e. 0.05 mg/ml). However, CPPms-III was more effective (p<0.01) in lowering iron-induced oxidation in liposomes at a concentration of 0.10 mg/ml. Both CPPms-I and -III were shown to possess the most antioxidative activity at the highest concentration, 0.50mg/ml, 64 inhibiting 66.4 and 53.4% of the oxidation respectively. Other known metal chelators such as phytic acid and E D T A (0.5mM) were also used in the Fe(II)-induced liposome oxidation as positive controls (Appendix; Figure A5) . Table 5. Rate of Propagation2 (% Control) of Liposome Peroxidation induced by Peroxyl Radical (AAPH) and % Inhibition3 of Lipsome Peroxidation induced by Transition Metal Ion (Fe 2 + )\ Assay CPPms Concentration of CPPms (mg/ml) 0.05 0.10 0.50 I 73.1 ± 1.4 a x 63.9 ± 0.7 a x 23.0 ± 6.8 b x A A P H 2 III 76.8 ± 1.8 a x 70.6 ± 3.3 a x 43.3 ± 2.6 b y I 11.4 ± 2.4 a x 11.6 ± 2 . 9 " 66.4 ± 10.8 b x ( F e 2 + ) 3 III 25.8 ± 1.3 a x 48.4 ± 2.0 b y 53.4 ± 2 . 1 b x 1. Results are expressed as Mean ± SD, n = 3 individual experiments; Means within the same row that do not share a common superscript letter (e.g. a, b and c) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (e.g. x and y) are significantly differently (p<0.01) 2. Rate of Propagation was determined by application of linear regression on kinetic graphs and expressed as % control. 3. % Inhibition [(1-A234nm(sampie/A234nm(controi))]*100 is calculated for individual time points and application of linear regression on kinetic graphs was used to achieve a constant (y-intercept) 65 3.2.5 Effect of C P P m s on scavenging stable D P P H and A B T S radicals The free radical scavenging activity of CPPms was determined by measuring the direct effect of CPPms on pre-formed stable radicals. The odd electron in the D P P H radical displays a strong absorption band at an absorbance of 517 nm, which is absent once the odd electron is paired off by a hydrogen- or electron-donating antioxidant. No significant hydrogen/electron donating activity was observed for either preparation of CPPms (Table 6), possibly due to the low solubility of CPPms in the solvent, ethanol, used in this assay. A more hydrophilic free radical, A B T S , was also used to evaluate the free radical quenching activity of CPPms. The reduction of monocation, A B T S + , accompanied by a reduction in absorbance at 734 nm, thus the extent of discoloration of A B T S + , when expressed as a percentage inhibition, is a function of the radical quenching affinity of CPPms. CPPms-I was shown to be a potent electron/proton donor with concentration dependent activity that stabilized up to 92.8% of the radicals at 1.0 mg/ml (Table 7). This activity was equivalent to the activity of 23.7 u,M Trolox. Hydrogen/electron donating activity was not as apparent in the CPP-III samples, which stabilized only 8.4% of the radicals at the highest concentration (1.0 mg/ml). 66 Table 6. Effect of CPPms-I and -III on Scavenging Stable DPPH Radicals.1 Concentration of CPPms (mg/ml) . CPPms 0.05 0.10 0.50 1.00 I -5.94 ± 4.8 a x -6.43 ± 2.28 a x 1.24 ± 1.20 a x 1.24 ± 4.60 a x III -7.10 ± 2.12 a x -4.87 + 1.08 a x -3.05 ± 1.02 a x 1.32 + 0.08 a x 1. Values represent % scavenging = Abs 5 1 9 n m m n t m i - Abs 519nm, a m p i f f xlOO; Abs 519nm c o ntroi Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (e.g. a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (e.g. x and y) are significantly differently (p<0.01). 67 Table 7. Effect of CPPms-I and -III on Scavenging Stable ABTS Radicals.1 Concentration of CPPms (mg/ml) CPPms 0.05 0.10 0.50 1.00 I 1 . 5 ± 0 . 7 a x 7.9 ± 1.0 b y 7 5 . 1 ± 1 . 6 c y 91.8 ± 0 . 5 d y (TE) 2 • (0.5) (2.2) (19.4) (23.7) III 0.2 ± 1.0 a x 1.1 ± 0 . 8 a x 3.2 ± 0.7 a x 8.4 ± 0.8 c x (TE) 2 (0.2) (0.4) (1.0) (2.3) 1. Values represent % scavenging = A734nm f :n ntrni - A 7 3 4 n m H a r j g xlOO; A734nm c o n t r oi Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (e.g. a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (e.g. x and y) are significantly differently (p<0.01) 2. T E = the calculated Trolox equivalence (LIM) using the standard curve equation y = 3.896x - 0.5844; where x stands for micromolars of Trolox and y for % scavenging. 68 3.3 Discussion As an in vitro antioxidant assessment, caseinophosphopeptides, CPPms-I and -III from Mei j i Seika Kaisha Ltd. (Japan), were analyzed for mineral sequestering and free radical quenching activities in both aqueous and emulsion models. Both samples of CPPms contained peptides with a molecular weight of less than 6kDa, which were consistent with the peptides that are derived from cc si-, aS2- and P-caseins reported by other researchers (Naito et al., 1990; Reynolds et al, 1994; Ono et al, 1998; Gagnaire et al, 1996). Other higher molecular weight peptides, derived from trypsin hydrolysis of casein, were found only in CPPms-I but not in CPPms-III. A qualitative analysis of metal sequestering activity, using a simple absorbance scan over the range of 200 to 600 nm, showed the interaction between CPPms and metal ions such as ferrous iron and ferric iron. Spectra for the CPPms-metal complexes exhibited distinctly different apparent absorbance peaks than the corresponding CPPms and metal ions. The metal sequestering affinity of C P P has been attributed to the presence of a high concentration of phosphoserine residues in close proximity, thereby creating a polar and anionic domain that is highly electrostatically favorable for cationic metal ions (Reeve and Latour, 1958). The observed affinity of phosphoseryl residues, present in a si-casein (64-68) and P-casein (15-19), towards iron was reported by Manson and Cannon (1978). Phosphoseryl residues catalyzed oxidation of iron from ferrous to the ferric state, thereby resulting in the formation of highly stable iron-phosphoprotein complexes. C P P can sequester up to 4 m M of ferrous iron, creating iron adducts through delicate conformational changes in the peptide and attachment of iron to the oxygen on 69 phosphoserine in a tetrahedral coordination structure (Gaucheron et al., 1996). Once the free ferrous iron is bound by C P P , iron can be oxidized but partial oxidation may yield 2 different species of CPP-iron complexes. This confirms the affinity of phosphoserines to form stable chelates with ferrous as well as ferric iron (Galdi and Valencia, 1988), as shown by the reported detection of both pCN( l -25) /Fe 3 + and P C N ( l -25) /Fe 2 + complexes when the p-casein derived C P P , pCN(l -25) , was incubated with ferrous iron (Bouhallab et al., 1991). Affinity of phosphoserine residues towards C a 2 + can be affected by changes in p H and ionic strength with a lower binding ability occurring at high ionic strength and p H lower than 7, while the affinity of phosphoserine residues towards F e 2 + was unaffected (Baumy and Brule, 1988). A p H of 7.0 and ionic concentration of 0.10 m M of phosphate was used in this study and provided an adequate sequestering environment for the peptides to interact effectively with the metal ions of interest. A n assay involving hydroxy] radical-induced degradation of deoxyribose, the backbone sugar found in deoxyribonucleic acid ( D N A ) in an ascorbate mediated Fenton reaction was used to investigate the potential antioxidant activity of C P P . Since deoxyribose has an affinity towards iron, the free ferrous iron, in the site-specific assay (e.g. F e 2 + + H2O2), interacts directly with deoxyribose before generation of hydroxyl radicals ( O H ) from Fenton reaction. With the addition of E D T A in the non site-specific assay (e.g. F e 2 + - E D T A + H202), direct F e 2 + binding to deoxyribose is restricted which favors the formation of ' O H in the aqueous medium. The antioxidant affinity of C P P is then directly related to its affinity for F e 2 + in the former assay, and to its scavenging capacity for ' O H radicals in the latter. 70 The lower hydroxyl radical scavenging activity in the presence of E D T A (non site-specific assay) for highly purified CPP-III samples indicates that antioxidant activity was comparatively more pronounced as a ferrous iron chelator, than as a hydroxyl radical quencher. In contrast, CPP-I was equally effective with either role. This observation is interesting in view of reports showing that a s i - and P-caseins can scavenge hydroxyl radicals in a deoxyribose model (Cervato et al., 1999). In addition to aqueous models, the antioxidant activity of C P P was also investigated in a phospholipid liposome (i.e. emulsion) model. Spherical liposomes were formed spontaneously when phospholipids (e.g. soybean lecithin) dispersed into an aqueous buffer. Oxidation was initiated by the addition of ferrous ions or free radicals, such as A A P H (2,2'-azobis-2-amidinopropane-dihydrochloride), and the extent of oxidation was monitored by the formation of conjugated diene products at an absorbance of 234 nm. A A P H is a hydrophilic radical initiator that produces carboxy radicals and peroxyl radicals in the presence of molecular oxygen. Trace amounts of metal ions were eliminated by chelax ion-exchange chromatography prior to the exposure to A A P H to eliminate the possibility that decomposition of l ipid peroxides was attributed to metal ion-catalyzed oxidation. Both CPP- I and -III were found to effectively lower the iron- and AAPH-induced oxidative damage of liposomes. The results provided strong evidence that that phosphopeptides are potent antioxidants not only in an aqueous model but also in a lipophilic model. Phosphate groups in both free and serine-bound forms (orthophosphate and phosphoserine) were shown previously to have a lower effect on oxidation, compared to the corresponding phosphorylated caseins (Cervato et al., 1999). Thus, it is logical to 71 conclude that phosphoserine groups in C P P may be capable of partitioning into emulsion droplets and available to lower oxidative damage triggered by free radicals and ferrous iron. C P P at the liposomal surface can donate hydrogen atoms to stabilize peroxyl radicals and therefore diminishing the peroxidative effect of radicals on the emulsion droplet. Thermolysis of azo-substances, such as A A P H , generates a continuous flux of peroxyl radicals (ROO') that react with nearby substrates, extracting hydrogen atoms from substrates such as lipids (L-H) to generate a new radical L ' [Equation 15]. Formation of this radical invokes a free radical chain reaction resulting in the deterioration of lipids. However, in the presence of a chain-breaking antioxidant (AH), such as C P P , the chain reaction is terminated [Equation 16] and oxidation inhibited. R O O ' + L - H ^ R O O H + L ' [15] R O O ' + A - H -> R O O H + A ' [16] Higher phosphopeptide content, as in CPP-III, were likely to be responsible for the two-fold greater A A P H quenching affinity at the liposome surface, when compared to CPP- I that has a lower phosphopeptide content. In addition to the free radical initiator, ferrous iron was also used as a pro-oxidant in this study because it has a high reactivity in emulsion models. Positively charged metal ions are attracted to the negatively charged droplet surface resulting in generation of more free radicals, at or near the droplet surface (Mei et al., 1999). It has been shown that F e 2 + associated much better with anionic rather than with cationic or non-ionic hexadecane emulsions (Mei et al., 1998). Oxidation thereby increases as iron-emulsion 72 droplet interactions increase. Iron-induced oxidation levels in salmon oil-in-water emulsions were shown to be higher at p H 7.0 than lower p H values, due to a lower solubility and a resultant precipitation of the metal onto the l ipid droplet surface. The net effect of this is the closer association of iron with the l ipid, which enhances oxidation rates (Mancuso et ah, 1999). Phosphoserines were shown to partition into the l ipid phase and successfully chelate iron donated by ligands, thereby significantly impeding oxidative deterioration of iron-fortified milk (Hegenauer el al., 1979). In the case of the positively charged phospholipid liposomes used in this study, the affinity of F e 2 + towards the emulsion droplet was apparent since the oxidation proceeded in an exponential manner. Negatively charged C P P may play a role in the repulsive electrostatic interactions occurring between charged emulsion droplets and charged oxidant by promoting CPP-metal association and lowering attachment of the metal onto the emulsion droplet. Metal chelators, such as phytic acid and E D T A , were chosen as positive controls to validate the efficacy of the ferrous iron-induced liposome oxidation. Both of these species were capable of lowering the extent of oxidation, as observed for different concentrations of C P P , confirming the previous findings that demonstrated the efficacy of E D T A and phytate (50-2000 LtM) in protecting SDS-salmon oi l emulsions from ferrous chloride (500 jxM) induced oxidation (Mei et al., 1998). Caseins have been shown to be most active in lowering T B A R S formation in Fe/ascorbate-induced peroxidation of arachidonic acid supplemented liposomes when compared to lactoferrin and lactoglobulin. Among the caseins, a-casein, carrying 10 phosphate groups, was shown to be most potent in extending the latency of peroxidation 73 when compared to P-casein with 5 phosphate groups and K-casein with only one phosphate group (Cervato et al, 1999). In addition to iron chelating activity, caseins also prevented hydroxyl radical formation, thereby suggesting that the mechanisms of antioxidant action were complex involving mechanisms other than simple iron-chelating activity. In a similar copper (Cu 2 + ) metal-induced oxidation model, caseins were more effective at protecting trilinolein emulsions than whole whey protein and a-lactalbumin (Allen and Wrieden, 1982). Other proteins derived from bovine milk, such as whey, are also active in lowering oxidation but through different mechanisms. Bovine whey proteins including iron-binding lactoferrin and serum albumin were shown to block transition metals and scavenge free radicals via aminio acid residues tyrosine and cysteine. With sulfhydryl residues, whey proteins scavenged peroxyl radicals and chelate iron to significantly lower formation of l ipid peroxides in salmon-in-oil emulsions (Tong et al., 2000), corn oi l emulsions (Huang et al., 1999), phosphatidylcholine liposomes (Colbert and Decker, 1991; Shinmoto et al, 1992; Heinonen et al, 1998; Wakabayashi et al, 1999) and infant formulas (Satue-Gracia et al, 2000). In order to verify the free radical quenching affinity of C P P , two additional free radical assays were used. The first assay involves a stable l,l-diphenyl-2-picrylhydrazyl (DPPH) radical that can accept an electron or hydrogen to become a stable, diamagnetic molecule invulnerable to oxidation. Because of its odd electron, D P P H shows a strong absorption band at 517 nm in ethanol (Blois, 1958). When this odd electron is paired-off by an antioxidant (AH) , the absorption vanishes and the resulting discoloration becomes a measure of the radical scavenging activity (Equation 17). 74 D P P H ' + A - H -> (DPPH) :H + A ' [17] However, C P P did not show any significant hydrogen/electron donating activity in this assay. This lack of activity could possibly be due to the low solubility of C P P in the solvent, ethanol, used. In a polar solvent, C P P structures can undergo conformation changes, possibly reducing the extent of contact between the peptide and the free radicals and rendering the acidic, hydrophilic functional domain ineffective. A second assay involves the radical monocations 2,2'-azinobis-(3-ethylenzothiazoline-6-sulfonic acid ( A B T S ' + ) that are generated by oxidation of A B T S with ammonium persulfate and reduced in the presence of hydrogen-donating antioxidants (Re et al, 1999). A similar hydrogen donation occurred in the A B T S assay: ( A B T S ) ' + + A - H -» ( A B T S ) : H + A ' [18] The addition of hydrogen/electron donating antioxidants (AH) to the monocation reduces it back to A B T S , thereby diminishing the strong absorption reading at 734 nm. Thus the extent of discoloration of A B T S + as percentage inhibition is a function of the radical quenching ability. Because of the higher hydrophilicity in the medium (i.e. deionized water), C P P was shown to effectively scavenge and hence stabilize the A B T S + cations. In contrast to the findings from the emulsion model, the crude CPP-I sample had a nine-fold more efficient hydrogen/electron donating than the CPP-III sample. The saturation of calcium ions, in the final steps of CPP-III preparation, may have interfered with the anionic functional domain created by the phosphoserines and contributed to a lowered activity to interact with A B T S + cations, despite the higher concentration of phosphoserine residues in the CPP-III sample. 75 In previous studies, this A B T S assay had been used extensively to detect antioxidant activity of various phenolic compounds in beverages and food extracts, but has yet to be applied to the milk system. However, free radical scavenging activity was identified in an isolated peptic hydrolysate of casein with a primary structure of Tyr-Phe-Tyr-Pro-Glu-Leu ( lkDa) (Suetsuna et al., 2000). This peptide also had a potent superoxide anion scavenging activity specifically at the Glu-Leu residues and also played an important role in quenching hydroxyl and D P P H radicals (Suetsuna et al., 2000). Although C P P does not contain this particular peptide, it is highly likely that the functional domain in C P P with the sequence SerP-SerP-SerP-Glu-Glu has comparable activity not only in metal chelation but also in hydrogen/electron donation and free radical scavenging. 3.4 Conclusion Caseinophosphopeptides (CPP) derived from tryptic digests of bovine casein, molecular weight 19 - 25kDa, were characterized to consist of molecular weights less than 6kDa. Two different C P P fractions, CPP-I and CPP-III, were shown to interact with ferrous iron, ferric iron and calcium ions and were screened for in vitro antioxidant activity. CPP-I and CPP-III exhibited significant (p<0.01) inhibition of both site-specific and non site-specific degradation of deoxyribose in a Fenton reaction oxidation test and protected liposomes from oxidation induced by ferrous iron and free radical A A P H . CPP-I was more effective (p<0.01) at quenching A B T S radicals than CPP-UI . These results demonstrate a potential affinity for bioactive casein derived peptides to prevent peroxidation reactions in labile food systems and reduce oxidative stress in the digestive tract. 76 3.5 Appendix 0.5 0.45 0.4 0.35 -I * 0 .3 -1 8 0.25 H 0.2 0.15 0.1 0.05 • E • H H 0 * 0 X X xx 20 40 60 80 Incubation time (mins) 100 120 Figure A l . Effect of CPP I at 0 mg/ml (•) , 0.05 mg/ml (•), 0.10 mg/ml (•) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by peroxyl radicals generated by addition of 2mM A A P H at 37°C. 77 20 40 60 80 Incubation time (mins) 100 120 Figure A2. Effect of CPP III at 0 mg/ml (•) , 0.05 mg/ml (•), 0.10 mg/ml (•) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by peroxyl radicals generated addition of 2mM A A P H at 37°C. 78 1.2 1 -0.8 -3 0.6 H < 0.4 0.2 • • • • . , • h b h h h • • • • • • x x x x x x x x x x x x x x X X X X x x X X 0 * 0 20 — I 1 1— 40 60 80 Incubation time (mins) 100 120 Figure A3. Effect of CPP I at 0 mg/ml (•) , 0.05 mg/ml (•), 0.10 mg/ml (•) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50 m M FeS0 4 at 37°C. 79 1.2 1 -0.8 -0.6 -0.4 0.2 • • • • • • • • • • • • • • • • • • • • • • • • • • • • •n n D • • u n V ' ^ ' v y x x X x x x x x x x x x x x • v v X x • I x x x X o * -20 40 60 80 Incubation time (mins) 100 120 Figure A4. Effect of CPP III at 0 mg/ml (•) , 0.05 mg/ml (•), 0.10 mg/ml (•) and 0.50 mg/ml (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50 m M FeS0 4 at 37°C. 80 1.2 1 -0.8 -3 0.6 0.4 0.2 0 * • • • • • • • • • • • • • • • • ° • • • • • • • • • • • • • _ • D • • X X X v X X v X v x x x x x x x x xx x xx x x T — I 1 1 1 0 20 40 60 80 100 120 Incubation time (mins) Figure A5. Effect of Control (•) , 0.5mM Phytic Acid (•) and 0.5 m M E D T A (X) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50 m M FeS04 at 37°C. 81 C H A P T E R IV: Antioxidant effects of laboratory prepared CPP and conjugates 4.1 Materials and Methods 4.1.1 CPP and conjugates formulations CPP-I , II and III were prepared by dissolving bovine milk casein in water making a 10% solution, with a p H adjusted to 7.5 to 8.5 (Kawano et al., 1991). Crystalline trypsin from porcine pancrease was added to a 0.01% (w/v) final concentration against the substrate and incubated at 50°C for 6 hours. CPP-I was obtained by spray-drying the whole hydrolysate. The hydrolysate was further purified by ion-exchange chromatography using Dowex 50wx8 cation exchange column (Dow Chemical Co.), followed by activated charcoal column (American Norit Co.) and was labeled CPP-II. CPP-IJI enriched with phosphopeptides was obtained by filtrating the insoluble mass from the hydrolysate at p H 4.5 before precipitating the phosphopeptides by adding calcium chloride (1.1% w/v) and ethanol (50% v/v) to the filtrate (Figure A12). C P P and galactomannan, a mannase hydrolysate of guar gum, (Taiyo Chemicals Co. Japan), or xyloglucan, a P-glucanase hydrolysate of tamarind seed (Dainihon Pharmaceuticals Osaka), were mixed in water at a weight ratio of 1:1 and lyophilized. The resulting powder mixture was incubated for 3 days at 60°C under 65% relative humidity. The incubated sample was applied on an open column (1.6 x 5.5cm) with DEAE-Toyopear l (Tosoh, Tokyo) equilibrated with 20 m M phosphate buffer (pH 7.4) and the column was washed with the same buffer to separate free carbohydrate. The C P P conjugates were eluted with 500 m M N a C l buffer (pH 7.4). Peak fractions were collected together and applied to an open column with Sephacryl S-300HR (Amersham/Pharmacia Biotech). The conjugates were eluted with 0 . 1 M phosphate 82 buffer (pH 7.4). Concentration of protein was directly determined by absorbance at 280 nm, while that of carbohydrate was determined by the absorbance at 470 nm after color development using phenol-sulfuric acid reaction. The molar binding ratio between C P P -III and galactomannan was 1:1.9 and for CPP-III and xyloglucan was 1:2.2. Calcium free samples of C P P conjugates were prepared by dissolving lOOmg C P P conjugates in 10 m M E D T A and stirring the mixture for 2 hours at room temperature before dialyzing against deionized water for 2 days at 4°C using Spectra/Por M W C O 1000 ( S P E C T R U M L A B Inc.). 4.1.2 Molecular Weight and Mineral Characterization Casein, CPP-I , CPP-II and CPP-III were analyzed by electrophoresis using 16.5% acrylamide Tris-Tricine (N-tris [hydroxymethyl] methyl glycine) SDS ready gel (10 well), according to the method of Laemmli (1970), and a Mini-Protean I M i n i - C e l l slab gel electrophoresis unit ( B I O - R A D Laboratories, Richmond, C A ) . A 4% acrylamide stacking gel, p H 6.8, was used. Protein samples (40Lig) and polypeptide molecular standards were heated at 100°C for 5 minutes in 1.0 M Tris-tricine buffer, p H 6.8 containing 10% SDS. Electrophoresis was carried out at a constant voltage of 100 volts for 100 minutes with a Tris-tricine running buffer. Gels were placed in 40% methanol and 10% acetic acid fixative solution for 30 minutes, stained in 0.025% Coomassie Blue G-250 solution (10% acetic acid) for 1 hour and destained in 10% acetic acid for 3 x 15 minute destain washes. Mineral characterization of C P P was performed on lOmg CPP-I , II and III by I C P - M S analysis for calcium, iron, magnesium, zinc and copper at Elemental Research Inc. (Appendix; Table A l ) . 83 4.1.3 Hydroxyl Radical Scavenging Assay in a Deoxyribose model A Fenton reaction model containing 0.1 m M of F e 3 + as the catalytic metal was used. The substrate 2-deoxyribose (3.6 m M ) was mixed together with C P P (0.05, 0.10, 0.50, 1.0 Omg/ml), 0.1 m M E D T A , 0.1 m M Ferric chloride, 0.1 m M ascorbic acid and I m M H2O2. The reaction mixture was incubated at 37°C for 1 hour. Following incubation, 1 ml of 10% (v/v) trichloroacetic acid ( T C A ) and 1% 2-thiobarbituric acid ( T B A ) were added and the mixture was boiled for 15 minutes at 95°C. Absorbance at 532 nm was recorded after cooling. The extent of deoxyribose degradation by hydroxyl radicals generated by the Fenton reaction was calculated using the equation: % Inhibition = Abs@532nmmnfrni - Abs@532nm H n 2 r j £ _x 100 Abs@532nm c o n t r o i 4.1.4 Peroxyl Radical Scavenging Assays in a Liposome model Liposomes were made by sonicating soybean lecithin (a-phosphatidylcholine) in 10 m M phosphate buffer (pH 7.4) in an ice-water bath for 20 minutes, according to Hu and Kitts (2000). Peroxyl radical induced-liposomal peroxidation was conducted at a constant temperature of 37°C using an A T I Unicam (UV2) U V / V i s spectrophotometer. The reaction was initiated by the addition of 0.2 m M A A P H to a mixture of 0.1 mg/ml of liposome in 10 m M phosphate buffer (pH 7.4) and C P P . Generation of conjugated diene hydroperoxide was monitored by taking absorbance readings at 234 nm every 4 minutes for 100 minutes. 4.1.5 Free radical scavenging assay The affinity of C P P to scavenge stable 2,2'-azinobis-3-ethylbenzothiazoline-6-sulfonic acid ( A B T S ) radical cations was investigated (Pellegrini et al, 1999). Distilled deionized water was used in place of ethanol for radical preparation. Discoloration was 84 determined by comparing the absorbance at 734 nm of the treatment groups with the control after an 8-minute incubation at room temperature. The inhibitory percentage of A B T S was calculated according to the following equation: % Inhibition = A b s @ 7 3 4 n i r w r n i - Abs@734nm , a m p i».x l00 Abs@734nm c o n t r o i 4.1.6 Statistical Analysis A l l data (with the exception of the liposome assays) were collected in triplicates and analysed by one way A N O V A (a < 0.01), followed by a multiple range Tukey post test analysis using the GraphPad Prism Analysis software (GraphPad Software Inc., San Diego C A ) to identify significant differences among treatment means (P < 0.01). Data were obtained from 3 different individual experiments for the liposome assays and submitted to linear regression using GraphPad software. Rates of propagation (i.e. slope of regression equation) were then analyzed by one way A N O V A (a < 0.01) and Tukey test (P < 0.01). 85 4.2 Results 4.2.1 Molecular weight and mineral content of C P P Using a tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis, T S D S - P A G E (Lesse et al, 1990), a higher resolution molecular weight characterization for C P P was achieved. It was found that CPP-I , II and III consisted of peptides of sizes that range from 3.5 to 6.5 kDa (Figure 11). Other researchers have reported that C P P contained peptides corresponding to the 2.6 kDa aS2-casein (2-21)-4P (Reynolds et al., 1994) or the 2.7 kDa a si-casein (59-79)-5P (Reynolds et al, 1994; Ono et al, 1998; Gerber and Jost, 1986). The higher molecular weight peptides present in CPP-I and II seemed to be larger hydrolysis fragments of casein proteins. Casein on the other hand had a range of peptides with molecular weights of 24 to 35 kDa, which seemed to correspond to the reported molecular weights of cc si, ocS2 and p\ around 22-23 kDa, 25 kDa and 23-34 kDa (Swaisgood, 1982). Each gram of CPP-I and II was shown to contain 400 to 500 p,g of calcium, whereas CPP-III had exceedingly high calcium content of 58.6 mg (Appendix; Table A l ) . A l l C P P preparations had a similar iron content of 200 to 400 jxg iron per gram peptide. The mineral profiles of C P P reflect the preparation methods employed. CPP-IJ had a relatively low calcium content, since the casein hydrolysates underwent cation-exchange chromatography as part of the purification procedures. In contrast, CPP-III was saturated with calcium ion because of the addition of calcium chloride in the final preparation steps. 86 Figure 11. 16.5% Tr i s -Tr ic ine S D S - P A G E Gel stained wi th Coomassie G-250 using the L a e m m l i buffer system (1.0 M Tr is , p H 8.45). Lane 1: 40 ug C P P - I , Lane 2: 40 ug C P P - I I , Lane 3: 40 ug C P P - I I I , Lane 4: 40 ug Casein, S td : Polypeptide Standards. 87 Phosphopeptide content in CPP-I , II and III were 12.3%, 12.3% and 86.1%, respectively (Saito et al., 1993). Phosphoserine concentration in CPP- I and CPP-III measured by H P L C was reported to be 0.16 and 1.22 mmol/g, respectively (Kasai et ah, 1995). H P L C analysis of CPP-III also confirmed the peptide content to be 83-93% (w/w), mainly consisting of residues (1-32) of bovine aS2-casein and residues (1-28) |3-casein (Hirayama et al., 1992b) and a nitrogen:phosphorus ratio of 8.0 (Saito et al., 1998). 4.2.2 Site-Specific and Non Site-Specific Scavenging Activity of CPP and Conjugates C P P inhibited the radical-induced degradation of deoxyribose by 2 processes, through metal chelating and hydroxyl radical scavenging. Oxidative degradation of deoxyribose was inhibited to a greater extent through free radical scavenging up to 60% at l.OOmg/ml C P P (Table 8; Figure 12a). In comparison, 32% inhibition occurred at the same concentration through ferrous iron chelation (Table 8; Figure 12b). C P P concentration was a greater factor in the metal chelating activity of C P P , than either the purity or calcium content. Although free radical scavenging activity also followed a concentration dependent increase, CPP-III was consistently more potent at inhibiting hydroxyl radical generation (Figure 12a), thus demonstrating the importance of the higher purity peptides in order to fully exploit C P P bioactivity. CPP-I , II and III were conjugated to polysaccharide galactomannan ( G A L ) and oligosaccharide xyloglucan ( X Y L ) , and the effect of glycosylation on the antioxidant activity of C P P was explored. Both saccharides are highly hydrophilic and upon attachment on the reducing ends of C P P result in an increase in surface activity and possibly altered structural conformation. 88 Hydroxyl radical scavenging activity and ferrous iron binding activity were both dramatically lower for the peptide-saccharide conjugates than for the peptides alone (Tables 9 and 10). Peptide-saccharide conjugates inhibited significantly (p<0.01) less deoxyribose oxidation than the corresponding peptide, at all concentrations tested in both assays. G A L - C P P conjugates inhibited, on average, 10-20% less deoxyribose oxidation than the corresponding C P P . In comparison, X Y L - C P P conjugates had at least 30% less inhibition of deoxyribose oxidation, or no inhibition at all. G A L alone resulted in 28.8% inhibition of deoxyribose oxidation, while X Y L exhibited 8.2% (Table 11; Table 13). It was found that G A L - C P P conjugates were not superior in antioxidant activity compared to simple, control mixtures of peptides and sugars. G A L and C P P mixtures were significantly more effective (p<0.01) in scavenging hydroxyl radicals at all concentrations tested, but had better (p<0.01) or equivalent (p>0.05) metal chelating activities than the G A L - C P P conjugates. X Y L - C P P conjugates had greater (p<0.01) radical scavenging activity, but lower ferrous iron binding activity (p<0.01) than corresponding mixtures. Removal of calcium via E D T A treatment following conjugation significantly lowered (p<0.01) the activity of the G A L - C P P conjugates regardless of concentration (Table 12), but improved (p<0.01) that of X Y L - C P P conjugates at concentration of 0.50 mg/ml and higher (Table 14). 89 Table 8. Effect of CPP I, II and III on Percent Inhibition of Degradation of Deoxyribose.1 Assay Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 Non Site-Specific Binding 2 CPPI CPPII CPPIII 25.3 ± 0.3 a x 29.9 ± 0.6 a y 34.7 ± 0.8 a z 34.2 ± 1.1 b x 34.9 ± 0.7 b x 50.7 ± 0.9 b y 53.6 ± 0.6 c x 52.9 ± 0.5 c x 60.0 ± 0.7 c y 61.3 ± 0.7 d x y 60.4 ± 0.7 d x 63.3 ± 0.7 d y CPPI 17.8 ± 0.5 a x 16.2 ± 0.9 a x 24.5 ± 0.5 b x 32.1 ± 1 .1 c x Site-26.5 ± 0.4 b x Specific CPPII 18.2 ± 0 . 1 a x 18.4 ± 0 . 1 a x y 32.9 ± 0.9 c x Binding 3 26.8 ± 0.3 b x CPPIII 16.8 ± 1.4 a x 19.5 ± 0 . 9 a y 32.1 ± 1.0 c x 1. % Inhibition = lOOx (A532nm c o n t r oi - A532nm s a m p i e ) / A532nm c o n troi; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01) 2. Ascorbic acid mediated Fenton Reaction in the presence of E D T A 3. Ascorbic acid mediated Fenton Reaction in the absence of E D T A 90 80 - i 70 H 1. % Inhibition = lOOx (A532nmCOntroi - A 5 3 2 n m s a m p i e ) / A532nm c o n t r oi; Results are expressed as Mean ± SD, n = 3; Reaction initiated by ascorbic acid mediated Fenton Reaction in the presence of E D T A (Non Site-Specific Binding Assay). 91 80 70 A 60 A g 50 1 40-c $ 30-20 A 10 0 0.05 0.10 0.50 1.00 Concentration of CPP (mg/ml) Figure 12b. Effect of CPP I (•), CPP II (•) and CPP III (El) on Percent Inhibition of Degradation of Deoxyribose.1 1. % Inhibition = lOOx (A532nmcontroi - A532nmsampie)/ A532nmcontroi; Results are expressed as Mean ± SD, n = 3; Reaction initiated by ascorbic acid mediated Fenton Reaction in the absence of EDTA (Site-Specific Binding Assay). 92 Table 9. Effect of CPP I, II and III and Conjugates on Percent Inhibition1 of Degradation of Deoxyribose2. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 CPPI 25.3 ± 0.3 a y 34.2 ± 1.1 b y 53.6 ± 0.6 c z 61.3 ± 0.7 d z Gal-CPPI 8.5 ± 0.9 a x 13.4 ± 0.4 b x 28.3 ± 0.5 c y 34.9 ± 0.8 d y Xyl-CPPI 11.8 ± 1.3 a x 11.8 ± 1.2 a x 20.5 ± 0.7 b x 26.3 ± 0.6 c x CPPII 29.9 ± 0.6 a y 34.9 ± 0.7 b z 52.9 ± 0.5 c z 60.4 ± 0.7 d z Gal-CPPII 13.0 ± 0.7 a x 21.3 ± 0.5 b y 32.9 ± 0.4 c y 36.6 ±1.0 d y Xyl-CPPII 12.4 ± 0.9 a x 16.2 ± 1.5 b x 23.4 ± 0.6 c x 27.2 ± 0.2 d x CPPIII 34.7 ± 0.8 a y 50.7 ± 0.9 b y 60.0 ± 0.7 c z 63.3 ± 0.7 * Gal-CPPIII 9.8 ± 0.8 a x 19.9 ± 0.5 b x 47.0 ± 1.1 c y 47.8 ± 0.8 c y Xyl-CPPIII 10.6+ 1.0 a x 16.8 ±0.9 b x 34.6 ± 0.4 c x 36.8 ± 1.1cx 1. % Inhibition = lOOx (A532nmcontroi - A532nmSampie)/A532nmControi; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of CPP and concentration that do not share a common superscript letter ( x, y and z) are significantly differently (p<0.01); Gal-CPP denotes conjugates of galactomannan and CPP, while Xyl-CPP denotes conjugates of xyloglucan and CPP. 2. Ascorbic acid mediated Fenton Reaction in the presence of EDTA (Non Site-Specific Binding Assay) 93 Table 10. Effect of CPP I, II and III and Conjugates on Percent Inhibition1 of Degradation of Deoxyribose . Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 CPPI 1 7 . 8 ± 0 . 5 a y 16.2 1 0.9 a y 24.5 10 .5 b z 32.11 1 .1 c z Gal-CPPI 0.3 ± 0.7 a x 3.1 11 .4 a x 7.91 1.9 b y 12.21 0.5 c y Xyl-CPPI 2.8 ± 1.5 b x 2.6 1 0.8 b x -5 .41 1.7 a x -6.1 1 1.8™ CPPII 18.210.1 a y 18.410.1 a y 26.5 1 0.4 b z 32.9 1 0.9 c z Gal-CPPII 7.1 1 1.5 a x 6 .511 .4 a x 9 .21 1.2 a y 14.41 1.7 b y Xyl-CPPII 2 .11 1.0 b x 2 .51 1.5 b x -6 .011.7 a x -4 .81 1.6 a x CPPIII 16.81 1.4 a y 19.51 0.9 a z 26.8 1 0.3 b z 32.11 1.0™ Gal-CPPIII 8.21 1.7 a x 10.0 1 0.5 a y 16.012.21 b y 19.6 ± 0.5 b y Xyl-CPPIII 5.31 1.7 b x 2.5 11 .6 a b x 0.2 1 2.0 a b x -0.4+ 1.2 a x 1. % Inhibition = 100x(A532nm c o n t r o i - A532nm s a m p i e ) /A532nm c o n t r oi ; Results are expressed as Mean 1 SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter ( x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while X y l - C P P denotes conjugates of xyloglucan and C P P . 2. Ascorbic acid mediated Fenton Reaction in the absence of E D T A (Site-Specific Binding Assay) 94 Table 11. Effect of Galactomannan and CPP I, II and III Conjugates and Mixtures on Percent Inhibition of Degradation of Deoxyribose.1 Assay Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 Gal 13.9 ± 0.7 a 21.6 ± 0.5 b 24.4 ± 1.3 b 28.8 ± 1.2 c Gal&CPPI 14.6 ± 0.3 a y 23.1 ±0.8 b y 42.3 ± 0.4 c y 50.0 ± 0.6 d y Non Site- Gal-CPPI 8 . 5 ± 0 . 9a x 13.4 ± 0.4 b x 28.3 ± 0.5 c x 34.9 ± 0.8 d x Specific Gal&CPPII 20.6 ± 0.8 a y 35.1 ±0 .2 b y 53.3 ± 1.0 c y 58.0 ± 1.1 d y Binding2 Gal-CPPII 13.0 ± 0.7 a x 21.3 ± 0.5 b x 32.9 ± 0.4 c x 36.6 ±1.0 d x Gal&CPPIII 15.4 ± 0.5 a y 33.8 ±0.7 b y 63.5 ± 0.4 c y 65.3 ± 0.6 c y Gal-CPPHI 9 . 8 ± 0 . 8 a x 19.9 ± 0.5 b x 47.0 ± 1.1cx 47.8 ± 0.8 c x Gal 5.6 ± 0.8 a 4.0 ± 1.0" 8.1 ±1 .4 a b 10.2 ± 1.5 b Gal&CPPI -3.8 ± 0.5 a x 5.8 ± 1.8 b x 7.6 ± 1.0 b x 13.8 ± 2.4 c x Site-Specific Gal-CPPI 0.3 ± 0.7 a x 3.1 ±1.4 a b x 7.9 ± 1.9 b c x 12.2 ± 0.5 c x Gal&CPPII 14.7 ± 1.3 a y 12.0 ±1.4 a b y 14.3 ± 1.5 a b y 17.6 ± 1.3 b x Binding3 Gal-CPPII 7.1 ± 1 . 5 " 6.5 ± 1.4" 9.2 ± 1.2" 14.4 ± 1.7 b x Gal&CPPIII 12.2 ± 0.8 a y 15.7 ± 0.5 a b y 19.1 ±0.7 b x 24.3 ± 0.6 c y Gal-CPPm 8.2 ± 1.7" 10.0 ± 0.5 a x 16.0 ± 2.21 b x 19.6 ± 0.5 b x 1. % Inhibition = 100x(A532nmcontroi - A532nmsampie)/ A532nmcontroi; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of CPP and concentration that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Gal-CPP denotes conjugates of galactomannan and CPP, while Gal&CPP denotes mixtures of galactomannan and CPP. 2. Ascorbic acid mediated Fenton Reaction in the presence of EDTA 3. Ascorbic acid mediated Fenton Reaction in the absence of EDTA 95 Table 12. Effect of Galactomannan and CPP I, II and III Conjugates and Calcium Free (CF) Conjugates on Percent Inhibition of Degradation of Deoxyribose.1 Assay Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 Gal-CPPI 8 . 5 ± 0 . 9 a x 13.410.4 b y 28.3 1 0.5 c y 34.9 10 .8 d y Non Site-Gal-CPPI-CF 5.0 ± 1.4 a x 8.011 .5 a x 16.2 1 0.6 b x 28.011 .8 c x Gal-CPPH 13.010.7 a y 21.310 .5 b y 32.9 1 0.4 c y 36.611.0 d y Specific Binding 2 Gal-CPPn-CF 3.91 1.2 a x 7.41 1.3 a x 17.81 1.1 b x 28.81 1.9 c x Gal-CPPHI 9 .81 0.8 a y 19.91 0.5 b y 47 .01 1.1 c y 47.8 1 0.8 c y Gal-CPPIII-CF 2.3 1 0.6 a x 3.9 1 0.4 a x 12.41 1.1 b x 26.011 .4 c x Gal-CPPI 0.3 1 0.7 a y 3.11 1.4 a b y 7 .91 1.9 b c y 12.21 0.5 c y Gal-CPPI-CF -34.611 .5 a x -36 .111.9™ -23.21 1.2 b x - 6 . 4 1 1 . 1 c x Site-Specific Binding 3 Gal-CPPH Gal-CPPn-CF 7.111.5 a y -36 .111 .4 a x 6.511 .4 a y -38.2 ± 1.2™ 9 .21 1.2 a y -24.31 1.6 b x 14.41 1.7 b y -5.811 .6 c x Gal-CPPm 8.21 1.7 a y 10.0 1 0.5 a y 16.0 1 2.21 b y 19.61 0.5 b y Gal-CPPIII-CF -44.61 1.4 a x -44.4 ± 1.7™ -31 .110.8 b x -12.71 1.8 c x 1. % Inhibition = 100x(A532nm c o ntroi - A532nm s a m pie)A532nmC Ontroi; Results are expressed as Mean 1 SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while G a l - C P P - C F denotes calcium free conjugates of galactomannan and C P P . 2. Ascorbic acid mediated Fenton Reaction in the presence of E D T A 3. Ascorbic acid mediated Fenton Reaction in the absence of E D T A 96 Table 13. Effect of Xyloglucan and CPP I, II and III Conjugates and Mixtures on Percent Inhibition of Degradation of Deoxyribose.1 Assay Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 Non Site-Specific Binding2 X y l 12.4 ± 1.7 b 7.9 ± 1.1 a b 5.2 ± 1.2 a 8.2 ± 1.5 a b X y l & C P P I X y l - C P P I 10.1 ± 0 . 2 a x 11.8 ± 1 . 3 " 11.8 ± 1 .4" 11.8± 1 . 2 " 20.9 ± 0.8 b x 20.5 ± 0.8 b x 27.6 ± 1.0 c x 26.3 ± 0.6 c x X y l & C P P I I X y l - C P P I I 6.1 ± 1 . 5 " 12.4 ± 0.9 a y 11.6 ± 1.2 b x 16.2 ± 1 . 5 b y 18.8 ± 0 . 1 c x 23.4 ± 0.6 c y 25.3 ± 0.8 d x 27.2 ± 0.2 d x X y l & C P P I I I Xyl -CPPII I 5.3 ± 1.3 a x 10.6 ± 1.0 a y 6.0 ± 0.5 a x 16.8 ± 0.9 b y 30.2 ± 0.7 b x 34.6 ± 0.4 c y 32.7 ± 1.0 b x 36.8 ± 1.1 c y Site-Specific Binding3 X y l -4.0 ± 2.3 b -10.3 ± 0.8 a -11.3 ± 0 . 8 a -11.5 ± 1 . 8 ' X y l & C P P I X y l - C P P I 12.2 ± 0.6 b y 2.8 ± 1.5 b x 13.9 ± 0.2 b y 2.6 ± 0.8 b x 7.6 ± 0.5 a y -5.4 ± 1 .7" 11.4 ± 0 . 2 b y -6.1 ± 1 . 8" X y l & C P P I I X y l - C P P I I 14.7 ± 1.3 c y 2.1 ± 1.0 b x 9.8 ± 1.2 a b y 2.5 ± 1.5 b x 6.9 ± 0.6 a y -6.0 ± 1 .7 " 10.6 ± 0.9 b y -4.8 ± 1 . 6 " X y l & C P P I I I Xyl -CPPI I I 8.2 ± 2 . 1 " 5.3 ± 1.7 b x 9.3 ± 1 . 8 a b y 2 . 5 ± 1 . 6 a b x 8.7 ± 0.6 a y 0.2 ± 2.0 a b x 12.6 ± 0.5 b y -0.4 ± 1 .2" 1. % Inhibition = 100x(A532nm c o n t r o i - A532nm s a m p i e )A532nm C ontroi; Results are expressed as Mean ± S D , n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); X y l - C P P denotes conjugates of xyloglucan and C P P , while X y l & C P P denotes mixtures of xyloglucan and C P P . 2. Ascorbic acid mediated Fenton Reaction in the presence of E D T A 3. Ascorbic acid mediated Fenton Reaction in the absence of E D T A 97 Table 14. Effect of Xyloglucan and CPP I, II and III Conjugates and Calcium Free (CF) Conjugates on Percent Inhibition of Degradation of Deoxyribose.1 Assay Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 Non Site-Specific Binding 2 X y l - C P P I X y l - C P P I - C F 11.8 ± 1.3™ 12.5 ± 1.0™ 11.8± 1.2™ 10.3 ± 1.6™ 20.5 ± 0 .8 b x 28.4 ± 0 .3 b y 26.3 ± 0.6C X 40.3 ± 0.3 c y X y l - C P P I I X y l - C P P I I - C F 12.4 ± 0.9™ 12.3 ± 0.4™ 16.2 ± 1.5b x 13.0 ± 1.2™ 23.4 ± 0.6C X 29.4 ± 1.6b y 27.2 ± 0 .2 d x 4 1 . 4 ± 0 . 4 c y Xyl-CPPI I I Xy l -CPPI I I -CF 10.6 ± 1.0™ 10.6 ± 0.9™ 1 6 . 8 ± 0 . 9 b y 9.2 ± 0.6™ 34.6 ± 0.4 c y 22.8 ± 0 .6 b x 36.8 ± 1.1" 35.1 ± 1.3CX Site-Specific Binding 3 X y l - C P P I X y l - C P P I - C F 2.8 ± 1.5b y -14.6 ± 2.8™ 2.6 ± 0 .8 b y -13.8 ±2.5™ -5.4 ±1.7™ 1 . 7 ± 0 . 4 b y -6.1 ± 1.8™ 1 8 . 7 ± 0 . 7 c y X y l - C P P I I X y l - C P P I I - C F 2.1 ± 1.0b y -15.9 ± 1.7™ 2.5 ± 1.5 b y -12.23 ± 1.4™ -6.0 ± 1.7™ 3.8 ± 0 .9 b y -4.8 ± 1.6™ 2 1 . 6 ± 0 . 5 c y Xyl-CPPII I Xy l -CPPI I I -CF 5.3 ± 1.7b y -18.5 ± 1.3™ 2 . 5 ± 1 . 6 a b y -14.5 ±2.0™ 0.2 ± 2 . 0 a b y -5.9 ± 1.5 b x -0.4 ± 1.2™ 5 . 6 ± 1 . 6 c y 1. % Inhibition = 100x(A532nm c o n t r oi - A532nm s a m p i e )A532nm C C ) n t r o i ; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter ( x, y and z) are significantly differently (p<0.01); X y l - C P P denotes conjugates of xyloglucan and C P P , while X y l - C P P - C F denotes calcium free conjugates of xyloglucan and C P P . 2. Ascorbic acid mediated Fenton Reaction in the presence of E D T A 3. Ascorbic acid mediated Fenton Reaction in the absence of E D T A 98 4.2.3 Effect of CPP and Conjugates on the formation of Conjugated Diene Addition of ferrous ion to the liposome emulsion system resulted in unexpected instability. Thus, kinetics for propagation varied and could not be reliably reproduced. Therefore, although trends were present for a lower oxidation (Figures A 6 , A 7 and A8) in the presence of C P P with the positive control, phytic acid ( ImM) , the variation in propagation kinetics of iron-induced oxidation in this assay did not allow for reliable calculation of propagation rates. In order to better demonstrate the bioactivity of C P P in scavenging free radical induced-oxidation by A A P H in an emulsion model, Trolox was added as a positive control at a concentration of 1.0 |xg/ml (Hu and Kitts, 2000). C P P was most potent at lowering peroxyl radical-induced damage on liposomes to 30% of control at the highest concentration (0.50 mg/ml) tested (Figures A 9 , A10 and A l l ) . This effect was comparable to the Trolox control (Table 15). Antioxidant activity of CPP-I , II and III followed a concentration dependent effect at lowering % control l ipid propagation rate and was not significantly different (p>0.05) among preparations at each concentration tested. Conjugation of CPP-I , II and III to G A L and X Y L did not affect C P P affinity to sequester peroxyl radical A A P H and lower liposome oxidation (Table 16). The presence of galactomannan at 0.50mg/ml significantly (p<0.01) lowered the propagation rate to 35.8%, having comparable effects as the Trolox control (Table 17). In contrast, xyloglucan at 0.50mg/ml did not have an effect at lowering propagation rate by maintaining 82.0 % control (Table 19). When the G A L and CPP- I or II were added to the emulsion as a mixture, the polysaccharide and the C P P when present as separate entities, lowered oxidation to a greater extent than when conjugated at concentrations of 99 0.10 mg/ml and 0.50 mg/ml, respectively (Table 17). A similar effect was observed for X Y L and CPP-I or II mixtures at higher concentrations (Table 19). However, mixtures of G A L or X Y L with CPP-III were not different from the conjugates in antioxidant activity. Calcium free conjugates of G A L and X Y L did not significantly improve calcium-containing conjugates activity (Table 18; Table 20). Table 15. Effect of CPP I, II and III on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Treatment Concentration of Trolox (Hg/ml) Concentration of CPP (mg/ml) 1.0 0.05 0.10 0.25 0.50 CPPI CPPII CPPIII 3 1 . 5 ± 3 . 3 a x 2 8 . 1 ± 2 . 3 a x 3 2 . 0 ± 1 . 3 a x 6 1 . 8 ± 2 . 3 b x 63.7±1.6 C X 67.7±4.3 b x 57 .0±3 .6 b x 5 6 . 9 ± 4 . 7 b c x 57 .6±3 .6 b x 4 5 . 6 ± 3 . 1 a b x 4 4 . 3 ± 4 . 0 a b x 4 6 . 3 ± 6 . 5 a b x 30 .6±1 .2 a x 33 .9±2 .7 a x 3 8 . 3 ± 6 . 8 a b x 1. Rate of Propagation (% Control) was c etermined by application of linear regression on kinetic graphs; Results are expressed as Mean ± SD, n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01). 100 Table 16. Effect of CPP I, II and III on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Treatment Concentration of Trolox (Lig/ml) Concentration of CPP (mg/ml) 1.0 0.05 0.10 0.25 0.50 CPPI 31.5 ± 3.3 a x 6 1 . 8 ± 2 . 3 b x 57.0+3.6 b y 45.613. l a b 30.611.2 a x G-CPPI 28.3 ± 2.5 a x 65 .2±7 .0 b x 53.5+8.7 b x y N D 22.312.3 a x X-CPPI 29.0 ± 1.0 a x 49.8±ll.lbx 33.417.2 a b x N D 24.211. l a x CPPII 28.1 ± 2.3 a x 63 .7±1 .6 C X 56.914.7 b c y 44.314.0 a b 33.812.7 a x G-CPPII 28.7 ± 5 .4 a x 66.6±10.3 b x 41.517.9 a b y N D 21.412.7a x X-CPPII 24.5 ± 0.5 a x 42.518.3 b x 25.112.6 a x N D 19.011.l a x CPPIII 32.0 ± 1.3 a x 67.714.3 b x 57.613.6 b y 46.316.5a b 38.316.8 a b x G-CPPIII 27.2 ± 7.8 a b x 70.517.0C X 49.317.0 b c x y N D 23.315.7a x X-CPPIII 28.1 ± 6.2 a x 73.218.5 c y 50.113.2 b c x N D 26.617.0a x 1. Rate of Propagation (% Control) was determined by application of linear regression on kinetic graphs; Results are expressed as Mean 1 SD, n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); N D represents values that were not determined; Ga l -CPP denotes conjugates of galactomannan and C P P , while X y l - C P P denotes conjugates of xyloglucan and C P P . 101 Table 17. Effect of Galactomannan and CPP I, II and III Conjugates and Mixtures on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Concentration of Concentration of Conjugate or Mixture Treatment Trolox (|j,g/ml) (mg/ml) 1.0 0.05 0.10 0.50 G 24.3 ± 2.4 a 63.5 ± 1.3 b 57.0 ± 1.0 b 35.8 ± 1.5 a G&CPPI 40.9 ± 5.0 a x 84.9 ± 3.7 c y 79.9 ± 4 . 6 c y 59.5 ± 3.8 b y G-CPPI 28.3 ± 2.5 a x 65.2 ± 7.0 b x 53.5 ± 8.7 b x 22.3 ± 2.3 a x G&CPPII 40.9 ± 5.0 a x 83.5 ± 4.8 b c x 75.8 ± 3.9 b y 55.0 ± 4.0 a b y G-CPPII 28.7 ± 5.4 a x 66.6 ± 10.3 b x 41.5 ± 7.9 a b x 21.4 ± 2.7 a x G&CPPIII 26.0 ± 1.7 a x 68.9 ± 0.4 b x 55.3 ± 1.5 b x 31.9 ± 0.5 a x G-CPPIII 27.2 ± 7.8 a x 70.5 ± 7 .0 c x 49.3 ± 7.0 b c x 23.3 ± 5.7 a x 1. Rate of Propagation (% Control) was determined by application of linear regression on kinetic graphs; Results are expressed as Mean ± SD, n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while G a l & C P P denotes mixtures of galactomannan and C P P . 102 Table 18. Effect of Galactomannan and CPP I, II and III Conjugates and Calcium Free (CF) Conjugates on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Treatment Concentration of Trolox (u.g/ml) Concentration of Conjugate (mg/ml) 1.0 0.05 0.10 0.50 G-CPPI 28.3+ 2.5 a x 65.2 ± 7 . 0 b x 53.5 ± 8.7 b x 22.3 ± 2.3 a x G-CPPI-CF 28.3 ± 2.5 a x 64.1 ± 10 .4 b x 58.0 ± 7.8 b x 27.6 ± 2.0 a x G-CPPII 28.7 ± 5.4 a x 66.6 ± 10.3 b x 41.5 ± 7.9 a b x 21.4 ± 2.7 a x G-CPPII-CF 24.0 ± 4.3 a x 57.4 ± 2.0 b x 45.1 ± 4.0 b x 22.2 ± 2.3 a x G-CPPHI 27.2 ± 7.8 a b x 70.5 ± 7.0 c x 49.3 ± 7.0 b c x 23.3 ± 5.7 a x G-CPPIII-CF 27.2 ± 7.8 a x 67.1 ± 8.2 b x 56.4 ± 6.5 b x 25.3 ± 6.4 a x 1. Rate of Propagation (% Control) was determined by application of linear regression on kinetic graphs; Results are expressed as Mean ± S D , n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while G a l - C P P - C F denotes calcium free conjugates of galactomannan and C P P . 103 Table 19. Effect of Xyloglucan and CPP I, II and III Conjugates and Mixtures on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Treatment Concentration of Trolox (ng/ml) Concentration of Conjugate or Mixture (mg/ml) 1.0 0.05 0.10 0.50 X 24.3 ± 2.4 a 95.7 ± 1.4 c 98.1 ± 3.6° 82.0 ± 7.5 b X & CPPI 26.0 ± 1.7™ 67.1 ± 3.7 b x 54.5 ± 3 . 4 b y 37.1 ± 0.9 a x X-CPPI 29.0 ± 1.0 a x 49.8 ± 11.1 b x 33.4 ± 7 . 2 a b x 24.2 ± 1.1 a x X & CPPII 27.4 ± 3.1 a x 71.1 ± 5.5 b y 65.1 ± 4.7 b y 39.3 ± 3.8 a y X-CPPII 24.5 ± 0.5 a x 42.5 ± 8.3 b x 25.1± 2.6 a x 19.0 ± 1.1 a x X&CPPIII 27.4 ± 3.1 a x 77.7 ± 3.4 b x 68.8 ± 4.8 b y 39.8 ± 3.0 a x X-CPPIII 28.1 ± 6.2 a b x 73.2 ± 8 .5 c x 50.1 ± 3 . 2 b c x 26.6 ± 7.0 a x 1. Rate of Propagation (% Control) was determined by application of linear regression on kinetic graphs; Results are expressed as Mean ± SD, n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); X y l - C P P denotes conjugates of xyloglucan and C P P , while X y l & C P P denotes mixtures of xyloglucan and C P P . 104 Table 20. Effect of Xyloglucan and CPP I, II and III Conjugates and Calcium Free Conjugates on the rate of propagation (% Control) of liposome peroxidation induced by peroxyl radicals (AAPH). 1 Treatment Concentration of Trolox (Lig/ml) Concentration of Conjugate (mg/ml) 1.0 0.05 0.10 0.50 X-CPPI 29.0+ 1.0 a x 49.8 ± 11.1 b x 33.4 ± 7.2 a b x 24.2 ± 1.1 a x X-CPPI-CF 29.0 ± 1.0 a b x 60.5 ± 6.6 b c x 49.1 ± 7.1 b x 26.8 ± 1.2 a x X-CPPII 24.5 ± 0.5 a x 42.5 ± 8.3 b x 25.1± 2.6 a x 19.0 ± 1.1 a x X-CPPII-CF 24.5 ± 0.5 a x 52.3 ± 5.7 b x 40.3 ± 4.6 b y 21.8 ± 0.8 a x X-CPPIII 28.1 ± 6.2 a b x 73.2 ± 8.5 c x 50.1 ± 3.2 b c x 26.6 ± 7.0 " X-CPPIII-CF 28.1 ± 6.2 a x 70.4 ± 7.5 b x 58.8 ± 6.1 b x 24.6 ± 4.3 a x 1. Rate of Propagation (% Control) was determined by application of linear regression on kinetic graphs; Results are expressed as Mean ± S D , n = 3 individual experiments; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); X y l - C P P denotes conjugates of xyloglucan and C P P while X y l - C P P - C F denotes calcium free conjugates of xyloglucan and C P P . 105 4.2.4 Effect of C P P and Conjugates on Scavenging A B T S radicals Assessment of C P P affinity to scavenge aqueous, stable free radical A B T S showed that CPP-TJ was the most pronounced scavenger, quenching up to 9 3 . 9 % of the radicals (equivalent to activity of 2 4 . 3 fiM Trolox) at 1 .00 mg/ml. CPP-III at 6 7 . 6 % scavenging (equivalent to 17 .5 / A M Trolox) and CPP-I were relatively less effective at stabilizing A B T S (Table 2 1 ; Figure 13) . This finding demonstrated the importance of C P P purification involving cation-exchange, for obtaining maximum performance of the functional phosphoseryl domain and hydrogen/electron donating activity. Conjugation of polysaccharide galactomannan and oligosaccharide xyloglucan improved the radical stabilizing activity of CPP-I , but lowered the activity of CPP-II. There was no effect at improving CPP-III activity (Table 2 2 ) . Radical scavenging of X Y L conjugates with CPP-I and II were significantly (p<0.01) greater than counterpart G A L - C P P - I and G A L - C P P - I I conjugates, respectively at 0 . 5 0 and 1.00 mg/ml. Galactomannan was capable of scavenging 8 6 . 8 % of the A B T S radicals, which is equivalent to the activity of 2 2 . 4 ixM Trolox, while xyloglucan did not further stabilize radicals to a significant extent (Table 2 3 ; Table 2 5 ) . A t concentrations of 0 . 5 0 mg/ml and higher, galactomannan conjugates with CPP-II and III were more potent hydrogen/electron donors, than corresponding mixtures. Regardless of this finding, the antioxidant activity for G A L - C P P - I were not different from simple mixtures of G A L and CPP-I . However, X Y L - C P P conjugates significantly (p<0 .01) improved A B T S stability than X Y L and C P P mixtures at all concentrations tested. Applying E D T A to G A L and X Y L conjugates to remove free calcium was shown to improve the functionality of the conjugates especially at higher concentrations (Table 2 4 ; Table 2 6 ) . The only exception 1 0 6 to this observation was the activity of calcium free X Y L - C P P - I I I conjugates, which had an activity that was equivalent to that of the calcium containing conjugates. Table 21. Effect of CPP I, II and III on scavenging stable ABTS radicals. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 CPPI 1 0.4 ± 2 . 6 a x 0.6 ± 2.2 a x 2.7± 0.7 a x 3.2 ± 1.0 a x (TE) 2 (0.3) (0.3) (0.9) (1.0) CPPII 1 1 . 0 ± 0 . 9 a x 4.7 ± 2 . 1 a x 76.9 ± 2.9 b y 93.9 ± 0.6 c y (TE) 2 (0.9) (1.4) (19.9) (24.3) CPPIII 1 2.3 ± 0.9 a x 4.5 ± 1.9 a x 45.7 ± 1.3 b z 67.6 ± 1.8 c z (TE) 2 (0.7) (1.3) (11.9) (17.5) 1. % Scavenging = lOOx (A734nm c o n t r o i - A734nm s a mpie)A734nm c o n t r o r , Results are expressed as Mean ± SD, n = 3 Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01) Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01) 2. T E denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (LIM) of Trolox and y for % scavenging. 107 Figure 13. Effect of CPP 1(B), CPP II (•) and CPP III (H) on scavenging stable ABTS radicals.1 1. % Scavenging = lOOx (A734nm c o n t r o i - A734nm s a mpie)A734nm C O ntroi; Results are expressed as Mean ± SD, n = 3 108 Table 22. Effect of CPP I, II and III and Conjugates on scavenging stable ABTS radicals. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 CPPI 1 (TE) 2 Gal-CPPI 1 (TE) 2 Xyl-CPPI 1 (TE) 2 0.4 ± 2.6 a x (0.3) 4.7 ± 1.1 a x y (1.3) 6.1 ± 1.4 a y (1.7 0.6 ± 2.2 a x (0.3) 5.0 ± 0.7 a x y (1.4) 8 . 6 ± 0 . 8 a y (2.4) 2.7± 0.7 a x (0.9) 32.2 ± 1.6 b y (8.4) 39.7 + 1.1 b z (10.3) 3.2 ± 1.0 a x (1.0) 67.1 ± 0 . 2 c y (17.4) 72.4 ± 1.0 c z (18.7) CPPII 1 (TE) Gal-CPPII 1 (TE) 2 Xyl-CPPII 1 (TE) 2 1 . 0 ± 0 . 9 a x (0.9) 4.9 ± 0.5 a x (1.4) 5.3 ± 1.1 a x (1.5) 4.7 ± 2 . 1 a x (1.4) 7.0 ± 0.5 a x y (1.9) 11.9 ± 0 . 4 b y (3.2) 76.9 ± 2.9 b z (19.9) 35.8 ± 0.7 b x (9.3) 50.3 ± 2.0 c y (13.1) 93.9 ± 0.6 c z (24.3) 70.7 ± 0.9 c x (18.3) 86.3 ± 3.9 d y (22.3) CPPIII 1 (TE) Gal-CPPIII 1 (TE) 2 Xyl-CPPIII 1 (TE) 2 2.3 ± 0.9 a x (0.7) 4.8 ± 0.7 a x (1.4) 5.8 +0.7 a x (1.7) 4.5 ± 1.9 a x (1.3) 5.5 ± 1.6 a x (1.6) 9.3 ± 0.2 a x (2.5) 45.7 ± 1.3 b x (11.9) 38.6 ± 0.5 b x (10.1) 37.6 ± 1.6 b x (9.8) 67.6 ± 1.8 c x (17.5) 71.9 ± 0.4 c x (18.6) 68.9 ± 2 . 1 c x (17.8) 1. % Scavenging = 100 x (A734nm c o n t r o i - A 7 3 4 n m s a m p i e /A734nm c o n t r o i ) ; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while X y l - C P P denotes conjugates of xyloglucan and C P P . 2. T E denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (uM) of Trolox and y for % scavenging. 109 Table 23. Effect of Galactomannann (G) and CPP I, II and III Conjugates and Mixtures on scavenging stable ABTS radicals. Concentration of CPP (mg/ml) Treatment 0.05 0.10 0.50 1.00 G 1 -0.5 ± 0.2 a 1.7 ± 0.3 a 59.2± 3.0 b 86.8 ± 1.1 c (TE) 2 (0.0) (0.6) (15.4) (22.4) G & CPPI 1 0.0 ± 0.4 a x -0.1 ± 0 . 6 " 2 9 . 1 ± 2 . 3 b x 67.6 ± 1.0 c x (TE) 2 (0.2) (0.1) (7.6) (17.5) G-CPPI 1 4.7 ± 1.1 a y 5.0 ± 0.7 a y 32.2 ± 1.6 b x 67.1 ± 0 . 2 c x (TE) 2 (1.3) (1.4) (8.4) (17.4) G & CPPII 1 2.0 ± 2.0 a x -1.6 ± 0 . 1 " 24.4 ± 1.7 b x 65.2 ± 1.8 c x (TE) 2 (0.7) (6.4) (6.4) (16.9) G-CPPII 1 4.9 ± 0.5 a x 7 . 0 ± 0 . 5 a y 35.8 ± 0 . 7 b y 70.7 ± 0.9 c y (TE) 2 (1.4) (1.9) (9.3) (18.3) G & CPPIII 1 - 1 . 0 ± 0 . 3 " 0.4 ± 1 . 5 " 17.4 ± 0.8 b x 24.0 ± 0.3 c x (TE) 2 (-0.1) (0.3) (4.6) (6.3) G-CPPIII 1 4.8 ± 0.7 a y 5.5 ± 1 . 6 a y 38.6 ± 0.5 b y 71.9 ± 0.4 c y (TE) 2 (1.4) (1.6) (10.1) (18.6) 1. % Scavenging = 100 x (A734nm c o ntroi - A734nmsampie)/A734nmControi; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration group that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while G a l & C P P denotes mixtures of galactomannan and C P P . 2. T E denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (LIM) of Trolox and y for % scavenging. 110 Table 24. Effect of Galactomannan and CPP I, II and III Conjugates and Calcium Free (CF) Conjugates on scavenging stable ABTS radicals. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 G-CPPI 1 (TE) 2 G-CPPI-CF 1 (TE) 2 4.7 ± 1.1 a x (1.3) 2.7 1 0.3 a x (0.9) 5 . 0 ± 0 . 7 a x (1.4) 6 . 8 ± 0 . 8 b x (1.9) 32 .211 .6 b x (8.4) 62.31 1.1 c y (16.1) 67.1 ± 0 . 2 c x (17.4) 86.8 1 0.3 d y (22.4) G-CPPII 1 (TE) 2 G-CPPII-CF 1 (TE) 2 4.9 ± 0.5 a y (1.4) 2.6 ± 0 . 1 a x (0.8) 7.0 ± 0.5 a x (1.9) 5.0 ± 0 . 3 b x (1.4) 35.8 1 0.7 b x (9.3) 62.2 1 0.8 c y (16.1) 70.7 1 0.9 c x (18.3) 87 .81 0.5 d y (22.7) G-CPPIII 1 (TE) 2 G-CPPIII-CF 1 (TE) 2 4.8 ± 0.7 a x (1.4) 3 . 9 ± 0 . 8 a x (1.1) 5 . 5 1 1 . 6 " (1.6) 9.9 10 .2 b y (2.7) 38 .610 .5 b x (10.1) 66.9 10 .4 c y (17.3) 71 .91 0.4 c x (18.6) 92.5 1 0.6 d y (23.9) 1. % Scavenging = 100x(A734nm c o n t r oi - A734nm s a m pie)A734nm C ontroi; Results are expressed as Mean 1 SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of C P P and concentration group that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Ga l -CPP denotes conjugates of galactomannan and C P P , while G a l - C P P - C F denotes calcium free conjugates of galactomannan and C P P . 2. T E denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (LIM) of Trolox and y for % scavenging. I l l Table 25. Effect of Xyloglucan (X) and CPP I, II and III Conjugates and Mixtures on scavenging stable ABTS radicals. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 X 1 (TE)2 1.0 ± 1.5 a (0.4) 0.3 ± 1.3 a (0.2) -0.2 ± 1.8" (0.1) 4.2 ± 0.7 a (1.2) X & CPPI1 (TE)2 X-CPPI 1 (TE)2 0.9 ± 0.6 a x (0.4) 6.1 ±1 .4 a y (1.7 - 1 . 0 ± 0 . 2 a x (-o.i) 8.6 ± 0.8 a y (2.4) 21.8 ± 0.4 b x (5.7) 39.7 ± 1.1 b y (10.3) 62.3 ± 1.2 c x (16.1) 72.4 ± 1.0 c y (18.7) X & CPPII1 (TE)2 X-CPPII1 (TE)2 0.9 ± 1.4 a x (0.4) 5.3 ± 1.1 a y (1.5) -1.1 ±0 .4 a x (-0.1) 11.9 ±0.4 a y (3-2) 25.4 ± 2.9 b x (6.7) 50.3 ± 2.0 b y (13.1) 63.9 ± 1.1 c x (16.5) 86.3 ± 3.9 c y (22.3) X & CPPIII1 (TE)2 X-CPPIII1 (TE)2 2.6 ± 0.5 a x (0.8) 5.8 ± 0.7 a y (1.7) 0.8 ± 0.2 a x (0.4) 9.3 ± 0.2 a y (2.5) 15.0 ± 0.9 a b x (4.0) 37.6 ± 1.6 b x (9.8) 34.4 ± 1.4 b x (9.0) 68.9 ±2.1 c y (17.8) 1. % Scavenging = 100 x (A734nmCOntroi - A734nmsampie)/A734nmcontroi; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column of CPP and concentration group that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); Xyl-CPP denotes conjugates of xyloglucan and CPP, while Xyl&CPP denotes mixtures of xyloglucan and CPP. 2. TE denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (uM) of Trolox and y for % scavenging. 112 Table 26. Effect of Xyloglucan and CPP I, II and III Conjugates and Calcium Free (CF) Conjugates on scavenging stable ABTS radicals. Treatment Concentration of CPP (mg/ml) 0.05 0.10 0.50 1.00 X-CPPI 1 (TE) 2 X-CPPI-CF 1 (TE) 2 6.1 ± 1.4 a x (1.7) 5 . 3 ± 0 . 1 a x (1.5) 8 . 6 ± 0 . 8 a x (2.4) 14.0 ± 0.7 b y (3.8) 39.7 ± 1.1 b x (10.3) 78.7± 1 .1 c y (20.3) 72.4 ± 1.0 c x (18.7) 95.8 ± 0.4 d y (24.7) X-CPPII1 (TE) 2 X-CPPII-CF1 (TE) 2 5.3 ± 1.1 a x (1.5) 4.9 ± 0.9 a x (1.4) 11.9 ± 0.4 a x (3.2) 14.1 ± 0 . 9 b x (3.8) 50.3 ± 2.0 b x (13.1) 75.5 ± 0 . 7 c y (19.5) 86.3 ± 3.9 c x (22.3) 94.5 ± 0 . 1 d y (24.4) X-CPPIII1 (TE) 2 X-CPPIII-CF1 (TE) 2 5.8 ± 0.7 a x (1.7) 3.2 ± 0.4 a x (1.0) 9.3 ± 0.2 a x (2.5) 5 . 5 ± 0 . 9 a x (1.6) 37.6 ± 1.6 b x (9.8) 44.9 ± 0.6 b x (11.7) 68.9 ± 2 . 1 c x (17.8) 75.9 ± 0.3 c x (19.6) 1. % Scavenging = 100 x (A734nm c o n t r o i - A734nm s a m p i e ) /A734nm c o n t r o l ; Results are expressed as Mean ± SD, n = 3; Means within the same row that do not share a common superscript letter (a, b, c and d) are significantly differently (p<0.01); Means within the same column that do not share a common superscript letter (x, y and z) are significantly differently (p<0.01); X y l - C P P denotes conjugates of xyloglucan and C P P , while X y l - C P P - C F denotes calcium free conjugates of xyloglucan and C P P . 2. T E denotes the calculated Trolox equivalence using the standard curve equation y = 3.896x - 0.5844, where x stands for micromolar (uM) of Trolox and y for % scavenging. 113 4.3 Discussion An antioxidant effect of different protein hydrolysates formed by enzymatic hydrolysis of food protein by-products, such as malt sprouts, brewer's grains and bovine erythrocytes has been shown to be significantly improved by heating them in the presence of glucose (Lingnert and Eriksson, 1980). Pure amino acid preparations in place of the peptides from the protein hydrolysates did not have the same effect at lowering linoleic acid oxidation, therefore implicating the importance of peptide chains over the amino acid composition in eliciting antioxidant potential. Milk P-casein was found to display higher surface activity and superior adhesion to oil-in-water emulsions than milk whey protein P-lactoglobulin and egg yolk phosvitin (Dickinson et al., 1991). Therefore it is logical that CPP, derived from casein, would also assume an amphiphilic nature, having versatile surface-active behavior attributed to peptide backbone flexibility and hydrophobic-hydrophilic balance. The effect of this physiochemical property on antioxidant activity was tested with CPP in this experiment. The purpose of this study was to determine the antioxidant effect of laboratory synthesized CPP, and whether conjugation with polysaccharides such as galactomannan or an oligosaccharide, such as xyloglucan, would alter the activity of CPP in both aqueous (water-water interface) and emulsion (oil-water interface) model systems. Conjugation of CPP with saccharides induces conformational change to peptide structure, possibly exposing the functional domains and altering interfacial adsorptivity and hydrophilicity. Preparation of CPP involved the hydrolysis of bovine casein by trypsin at the carboxyl ends of L-arginine or L-lysine on the protein, freeing lysine groups on CPP 114 which in turn serve as attachment sites for the reducing ends of galactomannan and xyloglucan during Mail lard reactions. C P P derived from a si-casein have 7 phosphate groups and 3 free amino groups, ideal as attachment sites for galactomannan and xyloglucan molecules. As for peptides derived from (3-casein with 4 phosphate groups, there is only 1 free amino group for sugar attachment. The polysaccharide galactomannan is composed of a (l->4)-p-D-mannan backbone, substituted with a side chain of a-D-galactose, linked (1-^6) to mannan residues (Grant Reid, 1985). The oligosaccharide xyloglucan, with a smaller molecular weight distribution range (hepta-, octa-, and nonamer), is derived from tamarind seed and composed of a (l->4)-P-D-glucan backbone substituted with a side chain of a-D-xylose and P-D-galactosyl-(l->2)-a-D-xylose linked ( 1 ^ 6 ) to glucose residues (Brinson and Dey, 1985). Steric hindrance that occurs during peptide conjugation to long chained galactomanan ( M W 14kDa) during Mail lard reactions keeps the reaction under control, whereas the lack of hindrance for short-chained xyloglucan ( M W 1.4kDa) conjugation may cause the Mail lard reaction to go much further. Three different preparations of C P P as well as conjugation products were used in the antioxidant characterization studies. CPP-I was a crude spray-dried product of whole bovine casein hydrolysis while CPP-II was derived from hydrolysates that were subjected to cation-exchange chromatography. CPP-III was enriched with phosphopeptides through filtration and precipitation via calcium chloride and ethanol, increasing the content of phosphopeptides from 12.3% in CPP-I and II to 86.1% in CPP-III (Saito et al, 1993). Mineral analyses of these peptides showed that CPP-III was highly saturated with calcium, whereas CPP-I and II contained relatively lower calcium contents. Using a 115 higher resolution Tris-tricine P A G E , it was shown that peptides with approximate molecular weights that correspond to aS2-casein (2-21), a si-casein (59-79) and P-casein (1-28) were identified in synthesized C P P fractions. A number of higher molecular weight peptides were also identified in CPP-I and II, suggesting the presence of other hydrolysis fragments of casein proteins. Table 27. Primary amino sequences of common CPP (Swaisgood, 1982). Casein Sequence asi-casein y / Ile*-Lys-Gln-Met*-Glu-Ala-Glu-SerP-Ile*- l SerP-5erP-5erP-(57-79) 4P G/w-G/a-I le*-Val*-Pro*-Asn-SerP-Val*-Glu-Gln-Lys 7 9 aS2-casein 2 Asn-Thr-Met*-Glu-His-Val*-SerP-5erP-SerP-G /M -G /M -SerP-(2-21) 5P Ile*-Ile*-SerP-Gln-Glu-Thr-Tyr*-Lys 2 1 P-casein 1 Arg-Glu-Leu*-Glu-Glu-Leu*-Asn-Val*-Pro*-Gly-Glu-I le*-(1-28) 4P Wa\*-G\u-SerP-Leu*-SerP-SerP-SerP-Glu-Glu-Ser-I\e*-Thr-Arg-Ue*-Asn-Lys hydrophobic residues The presence of a series of phosphoserine residues followed by 2 glutamate residues is common among all caseinophosphopeptides, and creates a highly electrostatically ideal domain for cationic metal binding (Table 28). The functionality of this domain is supported by the conformational structure of the various peptides. With a relative high content of phosphoseryl residues in the vicinity of hydrophobic residues such as valine, isoleucine, leucine and proline resides, the C P P can develop a distinct amphipathic nature with both polar and hydrophobic domains organized in a tertiary structure. Secondary structures around sites of phosphorylation were predicted to assume a oc-helix-p-loop-a-helix motif with the heavily phosphorylated domain in the loop region (Holt et al., 1989). C P P is also known to have a highly flexible backbone allowing residues remote from phosphorylated motif to also interact with metal ions (Cross et al, 2001). 116 The bioactivity of this functional domain can be challenged using different experimental conditions. For example, the potential antioxidant activity of C P P and associated conjugates in aqueous conditions was determined using an ascorbate mediated Fenton reaction that resulted in hydroxyl radical-induced degradation of deoxyribose. The presence of E D T A ( F e 2 + - E D T A + H2O2) favors oxidation of ferrous iron and the generation of hydroxyl radicals ( O H ) in the non site-specific binding assay. In the site-specific binding assay, E D T A is absent, allowing ferrous iron (Fe 2 + + H2O2) to interact directly with deoxyribose before proceeding to generate hydroxyl radicals from Fenton reaction. These two assays were used to determine the affinity of C P P to scavenge O H radicals and to chelate F e 2 + ' respectively. It was found that not only did C P P successfully inhibit oxidative damage of deoxyribose through chelating catalytic metals to reduce initiation reactions as predicted, but C P P was also shown to stabilize hydroxyl radicals, thus terminating oxidation. C P P -III was consistently stronger than CPP-I and II in hydroxyl radical scavenging activity. On the other hand, the different C P P preparations were not significantly different in relative affinity for ferrous iron. This result suggested that prior saturation with calcium, as in CPP-III, would not affect iron binding activity, simply because C P P had a much stronger affinity towards iron relative to calcium. C P P bound calcium ions would therefore be readily exchanged for iron in the aqueous system. The eight-fold higher phosphoserine content in CPP-III participated in greater hydroxyl radical quenching but did not make a difference in metal binding activity. Perhaps the majority of ferrous iron in the Fenton reaction was already bound by the 12% phosphoserine as in CPP-I and II, and therefore any further increase in phosphoserine 117 residues would not lower deoxyribose oxidation to a significant extent. A greater phosphoserine concentration, however, gave the CPP-III peptides a greater potential in electron donating activity to stabilize hydroxyl radicals, thus minimizing initiation and terminating the chain oxidation reactions. It was proposed that saccharides, being hydrophilic substances, would upon conjugation with C P P improve solubility of the peptides and enhance antioxidant activity in an aqueous model. In spite of an enhanced hydrophilicity, conjugation did not improve the affinity for C P P to exert antioxidant activity by quenching hydroxyl radicals, or binding to metal iron. In contrast, the polysaccharide and oligosaccharides actually reduced the natural antioxidant activity of C P P to different extents. Among the conjugates tested, C P P - G A L conjugates retained C P P antioxidant activity to a greater extent than C P P - X Y L conjugates, suggesting that saccharide chain length and possibly extent of the Mail lard conjugation might have a role in determining antioxidant activity. It has been shown that xyloglucan, an oligsaccharide derived from tamarind seeds, is highly prone to hydroxl radical attack (Miller and Fry, 2001), increasing the vulnerability of X Y L conjugates to free radical damage and significantly deterring antioxidant capacity. Addition of saccharides to the C P P backbone may have altered the configuration of phosphoseryl residues by disrupting the (3-loop groove of the phosphorylated sections of the peptide and thus exposing and rendering the reactive domain ineffective. The distortion in C P P conformation by glycosylation may contribute to the observed reduction in antioxidant activity of CPP-sugar conjugates, when compared to C P P alone. 118 In addition to the Fenton reaction model, an additional aqueous model was used to confirm the radical scavenging activity of C P P and conjugates. Primary antioxidant activity in C P P and conjugates was measured by the extent of reducing radical monocations ( A B T S + ) by hydrogen/electron donating activity. This assay seemed to be more sensitive with regards to the different relative compositions of C P P . While the crude CPP-I samples had the poorest electron donating power, the calcium-saturated, but higher purity CPP-III, exhibited moderate activity. The calcium free CPP-II proved the strongest (p<0.01) in stabilizing effect of A B T S + free radical cations. It was apparent that the presence of calcium cations on C P P seemed to obstruct the electron donating activity of phosphoserine residues. The stable A B T S + radicals in this assay are positively charged and electrostatically similar to calcium (Ca 2 + ) , thereby causing a possible charge repulsion at the functional domain which lowers activity to quench free radicals. Conjugation of C P P with galactomannan and xyloglucan lowered the activity of the most potent scavenger CPP-II , did not affect CPP-III scavenging activity, while slightly improved the activity of the weakest scavenger, CPP-I . Similar to the C P P , removing calcium from the conjugates seemed to significantly enhance antioxidant functionality. In comparison to the G A L - C P P - I and G A L - C P P - I I conjugates, the X Y L -C P P conjugates were consistently more effective at stabilizing the free radical through ^electron donation, possibly due to a lower steric hindrance amidst the phosphoserine-A B T S interaction. While glycosylation did not improve or affect the activity of more purified samples (i.e. CPP-II and III), it did help to improve in the electron donating activity of CPPI from 3.2% scavenging for CPP-I alone to 67.1% and 72.4% scavenging, for G A L - C P P - I and X Y L - C P P - I conjugates respectively. The mechanism by which 119 polysaccharide and oligosaccharide attachment improved the activity of CPP-I is not known at this time, however it is interesting that the crudest C P P samples showed comparable radical scavenging activity to the more purified C P P through conjugation with saccharides that had different chain lengths. Aside from aqueous models, the radical scavenging activity of C P P and conjugates in an emulsion model was also determined. Thermolysis of azo substances such as A A P H generates a continuous flux of peroxyl radicals ( R O O ) . Soybean phosphatidylcholine liposomes were shown to have peroxidizable membranes (Porter et al., 1980) that accelerated the peroxidation reactions of A A P H derived radicals generated in the membrane phase (Dean et al., 1991). The extent of antioxidant activity is limited by the chain-breaking affinity of C P P and associated conjugates to localize at the liposome membrane surface (i.e. oil-water interface), the site of radical generated oxidation. Linkage between surface-active peptides and saccharides could result in stabilization of the hydrophilic saccharide at the portion of the aqueous phase, while hydrophobic peptide residues are anchored in the oil droplets (Shu et al., 1996). The change in surface activity should place the peptide-saccharide immediately adjacent to the emulsion, where free radical oxidations occur (Frankel et al., 1994; Shepherd et al., 2000). It was shown that peroxyl radical scavenging activity was not different among the different preparations of C P P , thereby indicating that phosphoserine and calcium content were not as important as other surface activity factors. The amphiphilic property of C P P provide an ideal conformation for localization at the oil-water interface where it could proficiently quench peroxyl radicals and significantly lower the rate of propagation 120 reactions to 30.6-38.3% control. Conjugation of CPP with highly hydrophilic polysaccharides and oligosacchardies did not significantly affect the affinity of CPP to localize at the liposome membrane and improve CPP activity. It has been previously shown that conjugation of galactomanan with phosvitin (an iron-chelating egg yolk phosphoglycoprotein of 35kDa) increased the antioxidant activity of phosvitin in both iron-induced oxidation in phosphatidylcholine liposomes at pH 7.0 (Lee et al., 2002) and linoleic acid models (Uma Maheswari et al., 1997), or a powdered celite model system (Nakamura et al., 1998). No studies to date have established the radical scavenging activity of phosvitin in emulsion models, though phosvitin was capable of scavenging superoxide radicals in an aqueous model (Nakamura et al., 1998). Enhancement in protein functionality, through glycosylation with galactomannan was shown for whole proteins such as ovalbumin and lysozyme (Nakamura and Kato, 2000). However, perhaps protein hydrolysates in the case of CPP, which have a much lower molecular weight, behaved differently as compared to whole globular proteins. It appeared from this study that glycosylation did not improve the bioactivity of CPP to an extent that can be considered worthwhile for the cumbersome modification reactions. The polar acidic domain, which is a characteristic requirement for metal chelating affinity and also free radical scavenging for CPP in both aqueous and emulsion conditions, does not require further modification to enhance antioxidant activity. 121 4.4 Conclusion Results of this experiment showed that the different preparation techniques to produce C P P yield different calcium concentrations and peptide sizes. It appears that all C P P peptides carry the required polar, acidic Ser-P-Ser-P-Ser-P-Glu-Glu domain for ideal conformation for metal binding. C P P III was shown to be the most effective at binding iron and lowering deoxyribose oxidation, while CPP-II was more effective at sequestering A B T S radicals. A l l of the C P P preparations were equally effective at quenching peroxyl radicals in the liposome peroxidation model. These observations showed that C P P is both a primary and also a secondary antioxidant. Conjugation of C P P with galactomannan and xyloglucan did not enhance the bioactivity of C P P as predicted. 122 4.5 Appendix Table A l . ICP-MS analysis of CPP for minerals1. CPP Calcium Iron Magnesium Zinc Copper CPP-I 400 300 37 20 1 cpp-n 500 200 59 30 <1 cpp-ni 58600 400 163 70 <1 'Results shown in micrograms per gram (ppm) ~i r 40 60 80 Incubation time (mins) 120 Figure A6. Effect of CPP-I at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml ( • ) and 0.25 mg/ml (X) and ImM Phytic Acid ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50mM FeS0 4 at 37°C. 123 1.2 l H 0.8 H 3 0.6 H 0.4 0.2 • * • • i 8 * x X x x x x x x x • • x>< v X x X • • • • • D • • • X x x X • X 0 * • 20 40 60 80 Incubation time (mins) 100 120 Figure A7. Effect of CPP-II at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml ( • ) and 0.25 mg/ml (X) and ImM Phytic Acid ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50mM FeS04 at 37°C. 124 1 . 2 0 . 8 H 8 0 . 6 H 0 . 4 H 0 . 2 H • 9 X • » 5 • Q g x X X . • S 9 • • B 9 • • • • • ! • B v x x X x x X 0 * -2 0 i i i 4 0 6 0 8 0 Incubation time (mins) 1 0 0 1 2 0 Figure A8. Effect of CPP-III at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml ( • ) and 0.25 mg/ml (X) and ImM Phytic Acid ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by Fenton reaction generated by addition of 0.50mM FeS04 at 37°C. 1 2 5 0.6 Incubation time (mins) Figure A9. Effect of CPP-I at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml (•), 0.25 mg/ml (X) and 0.50 mg/ml ( A ) and 1.0 ug/ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2mM A A P H at 37°C. 126 Incubation time (mins) Figure A10. Effect of CPP-II at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml (•) , 0.25 mg/ml (X) and 0.50 mg/ml (A) and 1.0 jig/ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2mM A A P H at 37°C. 127 0.6 0.5 0.4 A 8 0.3 0.2 A 0.1 A * i x A X X X * A A i .•; 8 » 8 *!! sttssi--0 * -20 40 60 80 Incubation time (mins) 100 120 Figure A l l . Effect of CPP-III at 0 mg/ml (•), 0.05 mg/ml (•), 0.10 mg/ml (•) , 0.25 mg/ml (X) and 0.50 mg/ml (A) and 1.0 ng/ml Trolox ( • ) on the time course of formation of conjugated diene in the oxidation of structured liposome mediated by addition of 2mM A A P H at 37°C. 128 1 Spray Dry CPPI Milk Casein I Trypsin Treatment I Heat Inactivation • Carobxypeptidase Treatment HC1 Preciptiation t Spray Dry CPPII Supernatant ppte CaCl 2 ,EtOH Supernatant Precipitate I Spray Dry I cppni Figure A13. Preparation of CPP. (By: Nakamura, S., Personal Communications) 129 C H A P T E R V: C Y T O P R O T E C T I V E E F F E C S O F CPP O N C O L O N C A R C I N O M A C E L L S ( C A C O 2) A N D INTESTINAL E M B R Y O N I C C E L L S (INT 407) 5.1 Materials and Methods 5.1.1 Cell Culture conditions Caco-2 cells (HTB-37; American Type Culture Collection, Rockvil le, M D ) , derived from human colon carcinoma, and Int-407 cells ( C C L - 6 ; A T C C ) , derived from a human embryonic intestinal epithelial cell line, were used to study the protective effect of caseinophosphopeptides (CPP) on intestinal cell iron (Fe 2 + ) toxicity. Cells were grown as monolayers at 37°C with 5% CO2 in Eagle's minimum essential medium E M E M , supplemented with 2 m M L-glutamine, 100 U/ml penicillin, 100 ng/ml streptomycin, (1.5 g/L sodium bicarbonate), 1% (v/v) essential amino acids and 10% (v/v) fetal calf serum (FCS). Culture media and supplements were purchased from Invitrogen Canada Inc. (Burlington, ON). Cells were spilt at a ratio of 1:3 when cells reached confluency and viability was assayed. 5.1.2 Cytotoxicity assay Caco-2 cells and Int-407 cells were cultured at a density of 5.0 x 10 4 cells/lOOul on a 96-well plate until a tight monolayer was formed (1 to 2 days following seeding). The cells were washed twice with phosphate buffered saline solution (PBS) and then re-incubated with Fetal Calf Serum (FCS) reduced E M E M (Invitrogen Canada Inc., Burlington, ON) for 24 hours at 37°C prior to the addition of ferrous sulfate. Concentrations of ferrous (II) sulfate (Fisher Scientific, Fairlawn, NJ) ranging from 0.01 to 200 m M were dissolved in FCS-reduced E M E M prior to addition to cell cultures. Smaller increments of iron concentration were used in subsequent experiments to 130 determine an effective concentration that reduced cell viability to 50% of control (EC50) following a 24-hour incubation at 37°C. Caco-2 cells and Int-407 cells were prepared similarly as described above, with an additional 1-hour incubation period containing CPP-III before the addition of ferrous sulfate. Viabil i ty of cells supplemented with 5, 10, 25, 50, 100, 250, 500 and 1000 Ltg CPP-III per ml of FCS-reduced E M E M prior to iron treatment were compared to cells treated with iron alone to monitor the effect of CPP-III on cell toxicity. Viabil i ty of cells treated with CPP-III only was also determined to rule out any possible cytotoxic effects of the peptide alone. Concentration of ferrous iron used to achieve a 50% viability in Caco-2 cells was determined to be 2.5 m M , compared to the 4.0 m M in the Int-407 cells (Appendix; Figures A10 and A l l ) . 5.1.3 Cell Viability Viabil i ty of cells was assayed by monitoring the inhibition of M T T (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide; Sigma Chemical Co. , St. Louis, M O ) reduction in the colorimetric Mitochondrial Tetrazollium (MTT) assay (Mosmann, 1983). Cells were first rinsed twice with P B S before the addition of 75LI1 M T T stock solution at a concentration of 5 mg/ml P B S . Viable cells with a functional mitochondrial enzyme cleave the M T T salt to a dark blue formazan product. Following a 4 hour incubation at 37°C, 100 |xl of acid isopropanol (0.04N HC1 in 2-propanol; Fisher Scientific, Fair Lawn NJ) was added to all wells and vortexed thoroughly to dissolve the dark blue crystals. Viable cell count correlated directly with the concentration of formazan determined at absorbance 570 nm and was compared against a control to indicate the effect of various treatments on cell survival. 131 5.1.4 Statistical Analysis A l l experiments were done in duplicates with six replicates per experiment. A l l data were analyzed by one way A N O V A (a < 0.01) followed by a multiple range Tukey post test analysis with the GraphPad Prism Analysis software (GraphPad Software Inc., San Diego C A ) to identify significant different among treatment means (P < 0.01). 132 5.2 Results 5.2.1 Effect of Iron on cells Reduction of cell viability to nearly 60% of control was achieved with the addition of FeS0 4 for Int-407 (4.0 mM) and Caco-2 cells (2.5 mM). These specific concentrations of ferrous iron were selected according to a near 50% mortality observed for both cell lines in preliminary studies (Figures 14 and 15). 5.2.2 Effect of CPP-III on cells Laboratory synthesized, phosphoserine-enriched CPP-III was added to the apical surface of human intestinal embryonic Int-407 and human colon carcinoma Caco-2 cell monolaters in serum-reduced media. Viability (% control) of the 2 cell lines was measured by the M T T assay. CPP-III was shown to have no significant cytotoxic effects on the undifferentiated embryonic cells maintaining 98.5 to 106.2% viability of Int-407 cells (Table 29). CPP-III did not reduce the viability of the human colon cancer Caco-2 cells (100.4-103.6%) at a concentration range of 5-250 |0,g/ml. This result contrast the effect of CPP-m at 500 and 1000 ug/ml that significantly (p<0.01) lowered the % of cell survival to 98.0 and 82.8% respectively (Table 30). 5.2.3 Effect of CPP-III on iron toxicity of ceils CPP-III was found to inhibit the damage of iron-induced cytotoxicity in a dose-response manner, successfully recovering Int-407 cell viability to 94.2 and 98.4% of control at 500 and 1000 ng/ml respectively (Table 28). The cytoprotective effect of CPP-III in the Caco-2 cells increased with peptide concentration, resulting in a gradual rise in viable cell numbers to 87.8 and 88.6% at 500 and 1000 ng/ml respectively (Table 29). 133 140 120 -\ T 100 >, 80 > 60 H 0 4 6 8 10 Concentration of FeS04 (mM) 12 Figure 14. Viability of Int-407 cells following a 24-hour incubation with FeSC>4 at 37°C. 1 1. Viabil i ty (% control) = Abs 570nm^m j cii1 x 100; Abs 570nm c o n troi Results are expressed as Mean ± SD, n = 6; 134 2 4 6 8 10 Concentration of FeS04 (mM) 12 Figure 15. Viability of Caco-2 cells following a 24-hour incubation with FeS0 4 at 37°C. 1 1. Viabil i ty (% control) = Abs 570nm s a m p iP, x 100; Abs 570nm c o n t r oi Results are expressed as Mean ± SD, n = 6; 135 Table 28. Effect of CPP-III on the viability of Int-407 cells following a 24-hour incubation with 4 m M FeS0 4 at 3 7 0 C \ Treatment (Hg/ml) CPP only CPP & Fe 2 + 0 100 60 .3±3.9 a 5 103 .9±5 .3 a y 6 6 . 9 ± 4 . 7 a b x 10 105 .5±5 .3 a y 6 9 . 4 ± 5 . 3 b x 25 100 .2±4 .5 a y 6 6 . 1 ± 5 . 1 a b x 50 100.4±5.2 a y 7 0 . 2 ± 5 . 1 b x 100 103.3±4.6 a y 78.1±6.4 C X 250 98 .5±2 .9 a y 89 .4±6 .9 d x 500 101.7±3.6 a x 9 4 . 2 ± 6 . 3 d e x 1000 106.2±4.0 a x 98.4+4.6 e x 1. Viabil i ty (% control) = A570nm 5 3 Jm rjp_ x 100; A570nm c o n t r oi Results are expressed as Mean ± SD, n = 6; Means within the same column that do not share a common superscript letter (e.g. a, b, c, d and e) are significantly different (p<0.01); Means within the same row that do not share a common superscript letter (e.g. x and y) are significantly different (p<0.01). 136 Table 29. Effect of CPP-III on the viability of Caco-2 cells following a 24-hour incubation with 2.5 m M FeS0 4at 3TC1. Treatment (Lig /ml) CPP only CPP & Fe 2 + 0 100 60.9 ± 3.4 a 5 103.6± 1.7cy 78.9 ± 4 . 5 b c d x 10 102.5± 1.6 b c y 76.5 ± 4 . 3 b c x 25 100.4± 2 .9 b c y 74.2 ± 4 .4 b x 50 101.9+3.6 b c y 75.5± 6 .2 b c x 100 1 0 1 . 3 ± 2 . 4 b c y 8 5 . 9 ± 4 . 3 c d e x 250 100.4± 2 . 1 b c y 84.6± 4 . 7 c d e x 500 98.0± 2 .5 b y 87.8± 4 . 8 d e x 1000 82.8 ± 2.7 a x 8 8 . 6 ± 5 . 1 e x 1. Viabil i ty (% control) = Abs 570nm j f !ni Ew 1 x 100; Abs 570nm c o n t r o i Results are expressed as Mean ± SD, n = 6; Means within the same column that do not share a common postscript letter (e.g. a, b, c, d and e) are significantly different (p<0.01); Means within the same row that do not share a common postscript letter (e.g. x and y) are significantly different (p<0.01). 137 5.3 Discussion The human intestine is constantly challenged by the presence of dietary oxidants, mutagens and oxygen species or related reactive oxygen metabolites that reside in the luminal contents. With the consumption of diets containing highly unsaturated fats, luminal accumulation of lipid hydroperoxides can be considerable, promoting oxygen radical generation and propagation and accelerating lipid peroxidation (Aw, 1999). Not withstanding this, subtoxic levels of lipid peroxides (10/nM) can potentially disrupt intestinal redox homeostasis and initiate an apoptotic cascade that cannot be reversed even when the redox balance is restored (Wang et al, 2000). In addition to preformed radicals, diet-derived minerals, such as iron, may also act as oxidants, catalyzing Fenton reaction-induced oxidations and altering the cellular oxidation-reduction (redox) balance at the absorptive surface of the intestines (Nunez et ah, 2001). Imposing severe oxidant stress can lead to cytotoxicity of human intestinal epithelial cells. Intestinal cells respond differently to the degree of oxidative stress. For example, at mild oxidant concentrations (<10/xmol/L), intestinal proliferative activity increases, whereas at high oxidant concentrations (10-50/imol/L), intestinal cells die by apoptosis (i.e. cell shrinkage). At even higher concentrations of oxidants (>100/i,mol/L), there is significant necrotic cell death (i.e. cell lysis) (Aw, 1999). Through the Fenton reaction, the prooxidant iron can initiate a series of chain reactions resulting in the generation of free radicals such as hydroxyl radicals at the site of the intestine, causing severe damage to tissues and biomolecules. These events, in turn, may lead to cell ageing and death. Human colon carcinoma cells (Caco-2) are a suitable model for the study of transport and metabolism of possible toxicological 138 substances, such as iron, because Caco-2 cells accumulate iron in a time- and concentration-dependent manner, specifically from the apical pole (Halleux and Schneider, 1994). Excessive intracellular accumulation would then lead to cytotoxicity (Rossi et al, 1996). However, the iron uptake from the apical surface into enterocytes and transport across the Caco-2 cells are affected both by the valency of the iron and the iron status of the cells. Previous workers have shown that the uptake of ferrous iron (Fe 2 + ) was 200 fold greater than that of the ferric form (Fe 3 + ) , and the iron uptake was significantly higher for cells grown in conditions of low iron concentrations (Alvarez-Hernandez et al, 1991). The greater affinity for ferrous uptake at the brush border membrane of cells suggests that the intake of iron at the apical pole first requires a reduction of ferric iron prior to entry (Halleux and Schneider, 1994). Iron accumulation in Caco-2 cells was shown to increase protein and D N A oxidative damage, resulting in the loss of cell viability (Nunez et al, 2001). Iron induced D N A damage of intestinal epithelial cells in vitro and in vivo occurred by incubation of mucosal cells with 800 JXM. FeSCu and 12 /xM H2O2 and oral supplementation of iron to normal and iron-deficient rats (Srigiridhar et al, 2001). While a single oral dose of 8 mg iron caused apoptotic damage to rat intestinal mucosal cells, the continuous administration of the same dose for a period of 15 days resulted in necrosis of the GI tract absorptive surface as evidenced by the reduction in microvill i for normal rats and a complete erosion of v i l l i in iron deficient rats. The oxidative factors involved were identified in these studies to be hydroxyl and methoxyl radicals in the luminal and mucosal contents of rats (Srigiridhar et al, 2001). Ferrous iron crosses the enterocyte by 2 processes, which include a paracellular passage or a membrane receptor-facilitated pathway that is followed by intracellular 139 shuttling to the basolateral surface. It is hypothesized that ferrous iron chelators serve to inhibit iron uptake (Han et al., 1995) and accumulation in enterocytes, while free radical quenchers target the hydroxyl radicals generated at the cell membrane. The collective effect of these processes is to reduce the probability for the initiation and propagation of cell damage by free radical chain reactions. In this study, CPP-III was shown to be an effective iron-binding and radical quenching agent with potential cytoprotective effects on intestinal cells upon the insult of iron-induced cytotoxicity. The human colon adenocarcinoma cells, Caco-2 cells, in addition to a non-carcinomic human intestinal embryonic cell line Int-407 were used to test this effect. Caco-2 cells are adherent epithelial cells originially isolated from primary colonic tumor cells in humans. Upon reaching confluence, Caco-2 cells express characteristics of enterocytic differentiation in monolayers, with typical asymmetric morphology and the presence of glycogen deposits and microvil l i . The Caco-2 cells are seeded with the basolateral membrane attached to the surface of the culture dish and the microvillus-studded apical membrane facing the media. Integrity of monolyaters was reinforced by tight junctions in the apical zone and demosomes along the cells (Dharmsathaphorn et al., 1984). In spite of its cancerous origin, Caco-2 cells appear to behave similarly to the normal in vivo counterpart particularly under the differentiated state (Pageot et al, 2000). Similar to the Caco-2 cells, the Int-407 cells are also adherent epithelial cells, which are grown in monolayers with the only difference being the non-carcinomic nature. Int-407 cells, derived from human normal embryonic intestinal tissues isolated from the jejunum and ileum of embryos, were chosen in this study as a control, normal cell line for cytotoxicity assays. 140 Cel l viability, monitored by the M T T assay, was used as an end-measure for the iron toxicity assays. The tetrazolium salt ( M T T , or 3,(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) was metabolized to a colored formazan salt by mitochondrial enzyme activity found only in living cells and the extent of coloration correlated with viable cell numbers. This assay measures metabolic events that can more accurately quantify l iving cells and is more sensitive than dye exclusion assays that tend to overestimate viability as some reproductively dead cells fail to take up the dye (Wilson, 1992). Reduced serum and serum-free culture systems were shown recently to be superior to fetal bovine serum-containing systems because of a lower incidence of protein interference (Halleux and Schneider, 1994; Shinmoto et al, 1992; Maurer, 1992). Replacement of serum-containing medium by a defined nutritive medium, could increase the reproducibility of experiments and avoid interactions between serum constituents, while sustaining viability, growth rate and normal structural/functional differentiation in Caco-2 cells (Jumarie and Malo , 1991). It was found that C P P inhibited ferrous sulphate-induced toxicity in both the Caco-2 and Int-407 cells. Through tight binding with the ferrous iron, C P P removed the ferrous iron in the media and prevented iron uptake and intracellular accumulation. C P P alone displayed no cytotoxic effect on the Int-407 cells and only slightly reduced the viability of Caco-2 cells at the 2 highest concentrations. This slight reduction in viability of the Caco-2 cells by C P P alone may also be attributed to the unusually high calcium content of these C P P samples. This finding is consistent with the observed reduction of colonic epithelial cell proliferation, particularly that of precancerous or cancerous cells, 141 by dietary calcium supplements in rats (Liu et al, 2001; L i et al, 1998), thereby indicating the potential for calcium-containing ligands such as C P P to inhibit colon carcinogenesis in vivo (Vifias-Salas et al, 1998; Kampman et al, 2000). A high concentration of calcium was shown to lower the cell proliferation effect of free bile acids or fatty acids by converting them to insoluble calcium salts (McSherry et al, 1989) or binding to colon cell surface calcium receptors that signals the cells to stop proliferation (Pazianas et al, 1995). A s the Caco-2 epithelial cells differentiate, they also express alkaline phosphatase activity (Jovani et al, 2001), with the majority of phosphatase activity located at the apical side of the brush border membrane (Hidalgo et al, 1989). Yeung and colleagues have found that through partial hydrolysis of casein peptide bonds, phophoserines residues become exposed and as a result vulnerable to attack by intestinal alkaline phosphatase (IAP). This effect leads to the release of bound iron or the inhibition of iron chelation, which facilitates iron absorption (Yeung et al, 2001). The C P P dephosphorylation by I A P and the corresponding release of iron from the CPP-Fe complex may have contributed to the relative lower recovery in viability of Caco-2 cells, as compared to the control Int-407 cells in the present study. In vivo studies have shown that C P P exists not only in intestinal digests (Hirayama et al, 1992a) of rats fed casein or CPP-III diets, but also in the cecum and colon (Kasai et al, 1995) and subsequently in the fecal matter of rats (Kasai et al, 1992). The possibility exists therefore that the CPP-Fe complex would also bypass hydrolysis from proteolytic enzyme activity in the lumen or by enteric bacteria. A t the apical pole of the brush border membrane, the CPP-Fe complex is likely not absorbed into the 142 intestinal cells since the complex is neither recognized by the iron receptor on the membrane surface, or alternatively is small enough to enter through tight junctions for paracellular absorption. It is possible therefore that once the iron is bound to CPP, it collectively bypasses all digestive and absorptive processes and is excreted in a manner similar to CPP alone. Though derived from bovine milk, CPP behaves quite differently from other bovine milk proteins, such as (3-lactoglobulin, a-lactalbumin and lactoferrin that have all been shown to bind to brush border membranes of Caco-2 cells (Bolte et al, 1998) and transport through Caco-2 cell monolayers through absorptive endocytosis, releasing its iron following lysosomal degradation (Caillard and Tome, 1995; Sanchez et al, 1996). Although the levels of oxidation products were not determined in this study, the dual functionality of CPP as a metal chelator and also a free-radical stabilizer provides strong evidence that there is significant potential for CPP to prevent iron-induced oxidative damage to intestinal tissues associated with excessive and prolonged iron supplementation in vivo. CPP may also participate in quenching other pre-formed radicals, such as lipid peroxides or oxygen species, which have dietary origin. Studies to date have not been able to quantify the exact levels of CPP generated from CPP- or casein-containing diets, or to identify possible inhibitors present in the diet that counteract the bioactivity of CPP. Therefore, further research is warranted to establish the role of CPP as an active antioxidant in the context of dietary antioxidative supplements, or a therapeutic agent for correcting iron overload. 143 5.4 Conclusion Using a reduced serum culture system, CPP-III was shown to be relatively non-toxic to cells and to successfully recover human intestinal embryonic Int-407 and colon carcinoma Caco-2 cells from ferrous sulphate-induced cytotoxicity. Alkaline phosphatase activity present on the apical surface of differentiated Caco-2 cells may have contributed to a comparatively lower recovery of cell viability from cytotoxicity as compared to the control Int-407 cells. It seemed likely that iron-bound C P P would not be taken up by the enterocytes, but rather exit the digestive tract while bypassing most of the hydrolysis activity of proteolytic enzymes and enteric bacteria. 144 G E N E R A L C O N C L U S I O N A series of experiments were conducted to evaluate the potential in vitro antioxidant activity of caseinophosphopeptides (CPP), a tryptic hydrolysis product of bovine milk casein. First of all, CPP-I and CPP-III from Mei j i Seika Kaisha Ltd . (Japan) were analyzed for mineral sequestering and free radical quenching activities in both aqueous and emulsion models. CPP-III had a higher purity than CPP-I , but both were shown to contain peptides lower than 6kDa. A qualitative analysis of metal scavenging activity showed the interaction between CPP-I and III with metal ions such as ferrous iron, ferric iron and calcium. C P P suppressed the oxidation of deoxribose through site-specific binding with the pro-oxidant iron and through non site-specific binding with hydroxyl radicals. CPP-I and III were equally (p>0.05) effective in inhibiting deoxyribose oxidation through iron chelation, but CPP-I exhibited significantly (p<0.01) higher hydroxyl radical scavenging activity than CPP-III, at 0.50 and l.OOmg/ml. In an emulsion model, C P P was shown to lower oxidation through scavenging A A P H radicals and chelating ferrous iron at the surface of phospholipid liposomes. The ferrous ion sequestering activities of CPP-I and III in an emulsion were not significantly different (p>0.05) at 0.50mg/ml. However, CPP-I had a 2 fold lower (p<0.01) rate of propagation (% control) than CPP-III. In an aqueous free radical assay, CPP-I exhibited a significant (p<0.01) 91.8% scavenging for hydrophilic A B T S radicals while CPP-III had 8.4%. Hence, commercially available C P P acted as a primary and also a secondary antioxidant by displaying both metal sequestering and free radical quenching activity in aqueous and emulsion models. 145 Secondly, the antioxidant activities of laboratory synthesized C P P , along with CPP-saccharide conjugates, were evaluated to determine the effect of glycosylation on C P P antioxidant activity. CPP-I , II and III were shown to consist of peptides with molecular weights of 3.5 to 6.5 kDa, which correspond to peptides reported by other studies (Reynolds et al, 1994; Gerber and Jost, 1986; Gagnaire et al, 1996). CPP-III had the highest calcium concentration (58.6mg), as compared to CPP- I and II that contained 400 and 500ug of calcium. C P P inhibited degradation of deoxyribose via 2 processes, through metal chelating and hydroxyl radical scavenging. Oxidative degradation of deoxyribose was inhibited to a greater extent through free radical scavenging up to 60% at l.OOmg/ml C P P , in comparison to inhibition of 32% at the same concentration through ferrous iron chelation. CPP-III was consistently more effective at inhibiting hydroxyl radical generation. Conjugation of C P P with polysaccharide galactomannan ( G A L ) or oligosaccharide xyloglucan ( X Y L ) was shown to negatively impact the hydroxyl radical scavenging activity and ferrous iron binding activity of CPP , thereby inhibiting significantly (p<0.01) less deoxyribose oxidation than the corresponding peptide. In an emulsion model, C P P lowered AAPH-induced damage on liposomes to 30% of control at the highest concentration (0.50 mg/ml) and did not differ significantly (p>0.05) among preparations at each concentration tested. This antioxidant effect was comparable to the Trolox control. Conjugation of CPP-I , II and III to G A L and X Y L did not affect (p>0.05) C P P affinity to sequester peroxyl radical A A P H and lower liposome oxidation. Assessment of C P P affinity to scavenge aqueous, stable free radical A B T S showed that CPP-II was the most pronounced scavenger, quenching up to 93.9% of the radicals at l.OOmg/ml. A t l.OOmg/ml, CPP-III scavenged 67.6% of the 146 ABTS radicals, while CPP-I only scavenged 3.2% of the radicals. Conjugation of polysaccharide galactomannan and oligosaccharide xyloglucan improved the radical stabilizing activity of CPP-I, but lowered the activity of CPP-II. There was no effect at improving CPP-III activity. Laboratory synthesized, phosphoserine-enriched CPP-III was added to the apical surface of human intestinal embryonic Int-407 and human colon carcinoma Caco-2 cell monolaters in serum-reduced media. CPP-III was shown to have no significant cytotoxic effects on Int-407 cells, maintaining 98.5 to 106.2% viability. CPP-III did not reduce the viability of Caco-2 cells (100.4-103.6%) at a 5-250 Lig /ml but significantly (p<0.01) lowered the % viability to 98.0 and 82.8% at 500 and 1000 Lig /ml CPP-III, respectively. CPP-III was found to inhibit the damage of iron-induced cytotoxicity, recovering Int-407 cell viability from 60% (control) to 94.2 and 98.4% of control at 500 and 1000Lig/ml respectively. Similarly in the Caco-2 cells, the cytoprotective effect of CPP-III induced a rise in viable cell numbers from 60% (control) to 87.8 and 88.6% at 500 and 1000Llg/ml, respectively. These results demonstrate the potential affinity for bovine milk derived phosphopeptides to serve as a natural antioxidant, through metal sequestering and free radical scavenging activities, to lower the oxidative deterioration of lipids in food systems and the oxidative stress-related injury to biomolecules in vivo, especially in the intestine. 147 BIBLIOGRAPHY Adamson, N.J. and Reynolds, E.C. Characterization of Multiple Phosphorylated Peptides Selectively Precipitated from a Pancreatic Casein Digest. Journal of Dairy Science 1995, 78, 2653-2659. Ait-Oukhatar, N., Bouhallab, S., Bureau, F., Arhan, P., Maubois, J.-L., Drosdowsky, M . and Bougie, D.L. Bioavailability of caseinophosphopeptide bound iron in young rat. Journal of Nutritional Biochemistry 1997, 8, 190-194. Ait-Oukhatar, N., Bouhallab, S., Bureau, F., Arhan, P., Maubois, J.-L. and Bougie, D.L. In vitro digestion of caseinophosphopeptide-iron complex. Journal of Dairy Research 2000, 67, 125-129. Allen, J.C. and Wrieden, W.L. Influence of milk proteins of lipid oxidation in aqueous emulsion I. Casein, whey protein and a-lactalbumin. Journal of Dairy Research 1982, 49, 239-248. Alvarez-Hernandez, X., Nichols, G.M. and Glass, J. Caco-2 cell line: a system for studying intestinal iron transport across epithelial cell monolayers. Biochimica et Biophysica Acta 1991,1070, 205-208. Arrendondo, M . , Orellana, A., Garate, M.A. and Nunez, M.T. Intracellular iron regulates iron absorption and IRP activity in intestinal epithelial (Caco-2) cells. American Journal of Physiology - Gastrointestinal and Liver Physiology 1997, 273, G275-G280. Ashida, K., Nakajima, T., Hirabayashi, M . , Saito, Y., Matsui, T., Yano, H. Effects of Dietary Casein Phosphopeptides and Calcium Levels on Eggshell Quality and Bone Status in Laying Hens. Animal Science and Technology (Japan) 1996, 67, 961-91A. Aw, T.Y. Molecular and cellular responses to oxidative stress and changes in oxidation-reduction imbalance in the intestine. American Journal of Clinical Nutrition 1999, 70, 557-565. Bangham, A.D., Standish, M . M . and Watkins, J.C. Diffusion of univalent ions across the lamellae of swollen phospholipids. Journal of Molecular Biology 1965,13, 238. Barton, J . C , Conrad, M.E. and Parmley, R.T. Calcium inhibition of inorganic iron absorption in rats. Gastroenterology 1983, 84, 90-101. Baumy, J.-J. and Brule, G. Effect of pH and ionic strength on the binding of bivalent cations to p-casein. Lait 1988, 68, 409-418. Baumy, J.-J., Guenot, P., Sinbandhit, S. and Brule, G. Study of calcium binding to phosphoserine residues of P-casein and its phosphopeptide (1-25) by 3 I P NMR. Journal of Dairy Research 1989, 56, 403-409. 148 Benito, P. and Miller, D. Iron absorption and bioavailability: an updated view. Nutrition Research 1998,18, 581-603. Bennett, T., Desmond, A., Harrington, M . , McDonagh, D., FitzGerald, R., Flynn, A. and Cashman, K.D. The effect of high intakes of casein and casein phosphopeptide on calcium absorption in the rat. British Journal of Nutrition 2000, 83, 673-680. Berrocal, R., Chanton, S., Juillerat, M.A., Pavillard, B., Scherz, J.C. and Jost, R. Tryptic phosphopeptides from whole casein. II. Physiochemical properties related to the solubilization of calcium. Journal of Dairy Research 1989, 56, 335-341. Beucher, S., Levenez, F., Yvon, M . and Corring, T. Effects of gastric digestive products from casein on C C K release by intestinal cells in rat. Journal of Nutritional Biochemistry 1994, 5, 578-584. Blois, M.S. Antioxidant determination by the use of a stable free radical. Nature 1958,181, 1199-1200. Bolte, G., Wolburg, H., Beuermann, K., Stocker, S. and Stern, M . Specific interaction of food proteins with apical membranes of the human intestinal cell lines Caco-2 and T84. Clinica Chimica Acta 1998, 270, 151-167. Bouhallab, S., Leonil, J., Maubois, J-L. Complexation du fer par le phosphopeptide (1-25) de la caseine P: action de l'alcalase et de la phosphatase acide. Lait 1991, 71, 435-443. Bouhallab, S., Ait-Oukhatar, N., Molle, D., Henry, G., Maubois, J.-L., Arhan, P. and Bougie, D.L. Sensitivity of P-casein phosphopeptide-iron complex to digestive enzymes in ligated segment of rat duodenum. Journal of Nutritional Biochemistry 1999,10, 723-727. Brinson, K. and Dey, P.M. Polysaccharides Containing Xylose, Arabinose, and Galactose in Higher Plants. In Biochemistry of Storage Carbohydrates in Green Plants, Dey, P.M., Dixon, R.A., Eds., Academic Press: London, 1985. Brommage, R., Juillerat, M.A. and Jost, R. Influence of casein phosphopeptides and lactulose on intestinal calcium absorption in adult female rats. Lait 1991, 71, 173-180. Bronner, F. Intestinal calcium absorption: Mechanisms and applications. Journal of Nutrition 1987,117, 1347-1352. Brule, G. and Fauquant, J. Interactions des proteines du lait et des oligoelements. Lait 1982, 62, 323-331. Caillard, I. and Tome, D. Transport of P-lactoglobulin and a-lactalbumin in enterocyte-like Caco-2 cells. Reproduction, nutrition, developpement 1995, 35, 179-188. Cervato, G., Cazzola, R. and Cestaro, B. Studies on the antioxidant activity of milk caseins. International Journal of Food Sciences and Nutrition 1999, 50, 291-296. 149 Chabance, B., Marteau, P., Rambaud, J . C , Migliore-Samour, D., Boynard, M . , Perrotin, P., Guillet, R., Jolles, P. and Fiat, A . M . Casein peptide release and passage to the blood in humans during digestion of milk or yogurt. Biochimie 1998, 80, 155-165. Clare, D.A. and Swaisgood, H.E. Bioactive Milk Peptides: A Prospectus. Journal of Dairy Science 2000, 83, 1187-1195. Colbert, L.B. and Decker, E.A. Antioxidant Activity of an Ultrafiltration Permeate from Acid Whey. Journal of Food Science 1991, 56, 1248-1250. Cook, J.D. and Monsen, E.R. Vitamin C, the common cold, and iron absorption. American Journal of Clinical Nutrition 1977, 30, 235-241. Cook, J.D., Morck, T.A. and Lynch, S.R. The inhibitory effect of soy products on nonheme iron absorption in man. American Journal of Clinical Nutrition 1981, 34, 2622-2629. Cook, J.D. and Reddy, M.B. Effect of ascorbic acid intake on nonheme-iron absorption from a complete diet. American Journal of Clinical Nutrition 2001, 73, 93-98. Cross, K.J., Huq, N.L., Bicknell, W. and Reynolds, E . C Cation-dependent structural features of (3-casein-(l-25). Biochemical Journal 2001, 356, 277-286. Cross, M.L. and Gill, H.S. Immunomodulatory properties of milk. British Journal of Nutrition 2000, 84, S81-S89. Daniel, H. , Vohwinkel, M . and Rehner, G. Effect of Casein and P-Casomorphins on Gastrointestinal Motility in Rats. Journal of Nutrition 1990,120, 252-257. Dean, R.T., Hunt, J.V., Grant, A.J., Yamamoto, Y. and Niki, E. Free radical damage to proteins: The influence of the relative localization of radical generation, antioxidants, and target proteins. Free Radical Biology and Medicine 1991,11, 161-168. Demott, B.J. and Dincer, B. Binding Added Iron to Various Milk Proteins. Journal of Dairy Science 1976, 59, 1557-1559. Derman, D.P., Ballot, D., Bothwell, T.H., MacFarlane, B.J., Baynes, R.D., MacPhail, A.P., Gillooly, M . , Bothwell, J.E., Bezwoda, W.R. and Mayet, F. Factors influencing the absorption of iron from soya-bean protein products. British Journal of Nutrition 1987, 57, 345-353. Dharmsathaphorn, K., McRoberts, J.A., Mandel, K.G. , Tisdale, L.D. and Masui, H. A human colonic tumor cell line that maintains vectorial electrolyte transport. American Journal of Physiology - Gastrointestinal and Liver Physiology 1984, 246, G204-G208. Dickinson, E. , Hunt, J.A. and Dalgleish, D.G. Competitive adsorption of phosvitin with milk proteins in oil-in-water emulsions. Food Hydrocolloids 1991, 4, 403-414. 150 Ekmekcioglu, C , Pomazal, K., Steffan, I., Schweiger, B. and Marktl, W. Calcium Transport from Mineral Waters Across Caco-2 Cells. Journal of Agricultural and Food Chemistry 1999, 47, 2594-2599. Ellegard, K.H. , Gammelgard-Larsen, C , Sorensen, E.S. and Fedosov, S. Process scale chromatographic isolation, characterization and identification of tryptic bioactive casein phosphopeptides. International Dairy Journal 1999, 9, 639-652. Emery, T. Iron Oxidation by Casein. Biochemical and Biophysical Research Communications 1992,182, 1047-1052. Erba, D., Ciappellano, S. and Testolin, G. Effect of caseinphosphopeptides on inhibition of calcium intestinal absorption due to phosphate. Nutrition Research 2001, 21, 649-656. Faa, G. and Crisponi, G. Iron chelating agents in clinical practice. Coordination Chemistry Reviews 1989,184, 291-310. Feng, F.Q., Xu, S.Y. and Wang, Z. [Molecular size and calcium-binding capacity of casein phosphopetides.] Food and Fermentation Industries (China) 1997, 23, 18-21. Ferraretto, A., Signorile, A., Fiorilli, A. and Tettamanti, G. CPP influence on calcium uptake by the tumoral cells of intestinal origin HT-29 and Caco-2 in culture, 44th Congress of the Italian Society for Biochemistry and Molecular Biology, 1999, Vol. 13, 95. Ferraretto, A., Signorile, A., Gravaghi, C , Fiorilli, A. and Tettamanti, G. Casein phosphopeptides-induced [Ca ++]i changes in individual tumor cells of intestinal origin HT-29 cultured in vitro, 45th Congress of The Italian Society for Biochemistry and Molecular Biology, 2000, Vol. 15, 149. Ferraretto, A., Signorile, A., Gravaghi, C , Fiorilli, A. and Tettamanti,G. Casein Phosphopeptides Influence Calcium Uptake by Cultured Human Intestinal HT-29 Tumor Cells. Journal of Nutrition 2001,131, 1655-1661. Fiat, A . M . , Migliore-Samour, D., Jolles, P., Drouet, L. , Sollier, C.B.D. and Caen, J. Biologically Active Peptides from Milk Proteins with Emphasis on Two Examples Concerning Antithrombotic and Immunomodulating Activities. Journal of Dairy Science 1993, 76, 301-310. FitzGerald, R J . Potential Uses of Caseinophosphopeptides. International Dairy Journal 1998, 8, 451-457. Flynn, A. and Cashman, K. Nutritional Aspects of Minerals in Bovine and Human Milks. In Advanced Dairy Chemistry Volume 3 — Lactose, water, salts and vitamins; Fox, P.F., Ed.; Chapman & Hall: London, 1997. Fox, P.F. Milk proteins as food ingredients. International Journal of Dairy Technology 2001, 54, 41-55. 151 Frankel, E.N., Huang, S.-W., Kanner, J. and German, J.B. Interfacial Phenomena in the Evaluation of Antioxidants: Bulk Oils vs Emulsions. Journal of Agricultural and Food Chemistry 1994, 42, 1054-1059. Gagnaire, V., Pierre, A., Molle, D., Leonil, J. Phosphopeptides interacting with colloidal calcium phosphate isolated by tryptic hydrolysis of bovine casein micelles. Journal of Dairy Research 1996, 63, 405-422. Galdi, M . and Valencia, M.E. Stability of Iron (III) Chelates of Nutritional Interest. Journal of Food Science 1988, 53, 1844-1847. Garcia, M.N., Flowers, C. and Cook, J.D. The Caco-2 cell culture system can be used as a model to study food iron availability. Journal of Nutrition 1996,126, 251-258. Gaucheron, F., Famelart, M.-H. and Le Graet, Y. Iron-supplemented caseins: preparation, physicochemical characterization and stability. Journal of Dairy Research 1996, 63, 233-243. Gerber, H.W. and Jost, R. Casein Phosphopeptides: Their Effect on Calcification of In Vitro Cultured Embryonic Rat Bone. Calcified Tissue International 1986, 38, 350-357. Gill, H.S., Doull, F., Rutherfurd, K.J. and Cross, M.L. Immunoregulatory peptides in bovine milk. British Journal of Nutrition 2000, 84, S111-S117. Gillooly, M . , Bothwell, T.H., Torrance, J.D., MacPhail, A.P., Derman, D.P., Bezwoda, W.R., Mills, W. and Charlton, R.W. The effects of organic acids, phytates and polyphenols on the absorption of iron from vegetables. British Journal of Nutrition 1983, 49, 331-342. Gordon, M.H. The mechanism of antioxidant action in vitro. In Food Antioxidants; Hudson, B.J.F., Ed.; Elsevier Science Publishing Co., Inc.: New York, 1990. Grant Reid, J.S. Galactomannans. In Biochemistry of Storage Carbohydrates in Green Plants, Dey, P.M., Dixon, R.A., Eds., Academic Press: London, 1985. Greenberger, N.J., Balcerzak, S.P. and Ackerman, G.A. Iron uptake by isolated intestinal brush borders: Changes induced by alterations in iron stores. Journal of Laboratory and Clinical Medicine 1969, 73, 711-721. Gutteridge, J.M.C. and Halliwell, B. Antioxidants in Nutrition, Health, and Disease, Oxford University Press Inc.: New York, 1994. Hallberg, L. , Brune, M . and Rossander, L. Effect of ascorbic acid on iron absorption from different types of meals. Studies with ascorbic-acid-rich foods and synthetic ascorbic acid give in different amounts with different meals. Human Nutrition: Applied Nutrition 1986, 40A, 97-113. 152 Hallberg, L. , Rossander, L. and Skanberg, A.-B. Phytates and the inhibitory effect of bran on iron absorption in man. American Journal of Clinical Nutrition 1987, 45, 988-996. Hallberg, L. , Brune, M . , Erlandsson, M . , Sandberg, A.-S. and Rossander, L. Calcium: effect of different amounts on nonheme- and hem-iron absorption in humans. American Journal of Clinical Nutrition 1991, 53, 112-119. Halleux, C. and Schneider, Y.-J. Iron Absorption by Caco 2 Cells Cultivated in Serum-Free Medium as in Vitro Model of the Human Intestinal Epithelial Barrier. Journal of Cellular Physiology 1994,158, 17-28. Halliwell, B. and Gutteridge, J.M.C. Oxygen toxicity, oxygen radicals, transition metals and disease. Biochemical Journal 1984, 219, 1-14. Halliwell, B., Gutteridge, J.M.C. and Aruoma, O.I. The deoxyribose method: a simple "test tube" assay for determination of rate constants for reactios of hydroxyl radicals. Analytical Biochemistry 1987,165, 215-219. Halliwell, B. Antioxidant Characterization: Methodology and Mechanism. Biochemical Pharmacology 1995, 49, 1341-1348. Halliwell, B., Aeschbach, R., Loliger, J. and Aruoma, O.I. The Characterization of Antioxidants. Food Chemistry and Toxicology 1995, 33, 601-617. Hambraeus, L. Importance of milk proteins in human nutrition: physiological aspects. In Milk Proteins '84 Proceedings of the International Congress on Milk Proteins; Galesloot,T.E.; Tinbergen,B.J., Eds.; Pudoc: Wageningen, 1985. Han, O., Failla, M.L. , Hill, A.D., Morris, E.R. and Smith, J.C. Reduction of Fe(III) Is Required for Uptake of Nonheme Iron by Caco-2 Cells. Journal of Nutrition 1995, 125, 1291-1299. Hansen, M . , Sandstrom, B. and Lonnerdal, B. The Effect of Casein Phosphopeptides on Zinc and Calcium Absorption from High Phytate Infant Diets Assessed in Rat Pups and Caco-2 Cells. Pediatric Research 1996, 40, 547-552. Hansen, M . , Sandstrom, B., Jensen, M . and S0rensen, S.S. Casein phosphopeptides improve zinc and calcium absorption from rice-based but not from whole-grain infant cereal. Journal of Pediatric Gastroenterology and Nutrition 1997a, 24, 56-62. Hansen, M . , Sandstrom, B., Jensen, M . and S0rensen, S.S. Effect of Casein Phosphopeptides on Zinc and Calcium Absorption from Bread Meals. Journal of Trace Elements in Medicine and Biology 1997b, 11, 143-149. Hata, L, Higashiyama, S. and Otani, H. Identification of a phosphopeptide in bovine asi-casein digest as a factor influencing proliferation and immunoglobulin production in lymphocyte cultures. Journal of Dairy Research 1998, 65, 569-578. 153 Hata, I., Ueda, J. and Otani, H. Immunostirnulatory action of a commercially available casein phosphopeptide preparation, CPP III, in cell cultures. Milchwissenschaft 1999, 54, 3-7. Hegenauer, J., Saltman, P., Ludwig, D., Ripley, L. and Ley, A. Iron-Supplemented Cow Milk. Identification and Spectral Properties of Iron Bound to Casein Micelles. Journal of Agricultural and Food Chemistry 1979, 27, 1294-1301. Heinonen, M . , Rein, D., Satue-Gracia, M.T., Huang, S.-W., German, J.B., and Frankel, E.N. Effect of Protein on the Antioxidant Activity of Phenolic Compounds in a Lecithin-Liposome Oxidation System. Journal of Agricultural and Food Chemistry 1998, 46, 917-922. Hidalgo, I.J., Raub, T.J. and Borchardt, R.T. Characterization of the Human Colon Carcinoma Cell Line (Caco-2) as a Model System for Intestinal Epithelial Permeability. Gastroenterology 1989, 96, 736-749. Hirayama, M . , Toyota, K., Hidaka, H. and Naito, H. Phosphopeptides in Rat Intestinal Digests after Ingesting Casein Phosphopeptides. Bioscience, Biotechnology and Biochemistry 1992a, 56, 1128-1129. Hirayama, M . , Toyota, K., Yamaguchi, G., Hidaka, H. and Hiroshi, N. HPLC Analysis of Commercial Casein Phosphopeptides (CPP). Bioscience, Biotechnology and Biochemistry 1992b , 56, 1126-1127. Holt, C , Van Kemenade, M.J., Nelson, L.S.Jr., Sawyer, L. , Harries, J.E., Bailey, R.T. and Hukins, D.W.L. Composition and structure of micellar calcium phosphate. Journal of Dairy Research 1989, 56, 411-416. Holt, C , Timmins, P.A., Errington, N. and Leaver, J. A core-shell model of calcium phosphate nanoclusters stabilized by beta-casein phosphopeptides, derived from sedimentation equilibrium and small-angle X-ray and neutron-scattering measurements. European Journal of Biochemistry 1998, 252, 73-78. Hu, C. and Kitts, D.D. Studies on the Antioxidant Activity of Echinacea Root Extract. Journal of Agricultural and Food Chemistry 2000, 48, 1466-1472. Huang, S.-W., Satue-Gracia, M.T., Frankel, E.N. and German, J.B. Effect of Lactoferrin on Oxidative Stability of Corn Oil Emulsions and Liposomes. Journal of Agricultural and Food Chemistry 1999, 47, 1356-1361. Hunt, J.R. How important is dietary iron bioavailability? American Journal of Clinical Nutrition 2001, 73, 3-4. Huq, N.L., Cross, K.J. and Reynolds, E.C. A " H - N M R study of the casein phosphopeptide ocsi-casein (59-79). Biochimica et Biophysica Acta 1995, 1247, 201-208. 154 Hurrell, R.F., Lynch, S.R., Trinidad, T.P., Dassenko, S.A. and Cook, J.D. Iron absorption in humans as influenced by bovine milk proteins. American Journal of Clinical Nutrition 1989, 49, 546-552. Jackson, L.S. and Lee, K. The Effect of Dairy Products on Iron Availability. Critical Reviews in Food Science and Nutrition 1992, 31, 259-270. Jovani, M . , Barbera, R., Farre, R. and de Aguilera E . M . Calcium, Iron, and Zinc Uptake from Digests of Infant Formulas by Caco-2 Cells. Journal of Agricultural and Food Chemistry 2001, 49, 3480-3485. Juillerat, M.A. , Baechler, R., Berrocal, R., Chanton, S., Scherz, J.C. and Jost, R. Tryptic phosphopeptides from whole casein. Journal of Dairy Research 1989, 56, 603-611. Jumarie, C. and Malo, C. Caco-2 Cells Cultured in Serum-Free Medium as a Model for the Study of Enterocytic Differentiation In Vitro. Journal of Cellular Physiology 1991, 149, 24-33. Kampman, E. , Slattery, M.L. , Caan, B. and Potter, J.D. Calcium, vitamin D, sunshine exposure, dairy products and colon cancer risk (United States). Cancer Causes and Control 2000,11, 459-466. Kasai, T., Honda, T. and Kiriyama, S. Caseinphosphopeptides (CPP) in Feces of Rats Fed Casein Diet. Bioscience, Biotechnology and Biochemistry 1992, 56, 1150-1151. Kasai, T., Iwasaki, R., Tanaka, M . and Kiriyama, S. Caseinphosphopeptides (CPP) in Feces and Contents in Digestive Tract of Rats Fed Casein and CPP Preparations. Bioscience, Biotechnology and Biochemistry 1995, 59, 26-30. Kato, A., Murata, K. and Kobayashi, K. Preparation and characterization of Ovalbumin-Dextran conjugate having excellent emulsifying properties. Journal of Agricultural and Food Chemistry 1988, 36, 421-425. Kato, A., Sasaki, Y., Furuta, R. and Kobayashi, K. Functional protein-polysaccharide conjugate prepared by controlled dry-heating of ovalbumin-dextran mixtures. Agricultural Biology and Chemistry 1990, 54, 107-112. Kawano, T., Otsuka, R., Hidaka, H. and Naito, H. Patent [S-59-159792], 483-486. 1991. Japan. Kelly, P.M. and McDonagh, D. Innovative Dairy Ingredients. The World of Food Ingredients 2000, Oct/Nov, 24-32. Kennefick, S. and Cashman, K.D. Investigation of an in vitro model for predicting the effect of food components on calcium availability from meals. International Journal of Food Sciences and Nutrition 2000, 51, 45-54. 155 Kihara, Y. and Otani, H. Preparation of casein phosphopeptides from different kinds of pasteurized milks and their action on proliferative responses of mouse spleen cells in cell cultures. Milk Science (Japan) 2000, 49, 73-79. Kim, M . , Lee, D.T. and Lee, Y.S. Iron absorption and intestinal solubility in rats are influenced by dietary proteins. Nutrition Research 1995,15, 1705-1716. Kinsella, J.E. Protein Modification: Effects on Functional Properties and Digestibility. In Milk proteins: nutritional, clinical, functional and technological aspects; Barth, C.A.; Schlimme, E. , Eds.; Steinkopff Verlag Darmstadt: Germany, 1989. Kitts, D.D. and Kwong, W. Calcium bioavailability from Dairy Proteins. In Functional Foods: Dairy Products; Shortt, C , Ed.; Technomic Publishing Co., Inc.: Lancaster, PA, 2002. Kitts, D.D., Leung, R. and Nakai, S. Extrinsic Labelling of Caseinophosphopeptides with 4 5Calcium and Recovering Following Thermal Treatment. Canadian Institute of Food Science and Technology 1991, 24, 278-282. Kitts, D.D. and Yuan, Y.V. Caseinophosphopeptides and calcium bioavailability. Trends in Food Science and Technology 1992, 3, 31-35. Kitts, D.D., Yuan, Y.V. , Nagasawa, T. and Moriyama, Y. Effect of casein, caseinophosphopeptides and calcium intake on ilea 4 5 C a disappearance and temporal systolic blood pressure in spontaneously hypertensive rats. British Journal of Nutrition 1992, 68, 765-781. Kitts, D.D. Bioactive substances in food: identification and potential uses. Canadian Journal of Physiology and Pharmacology 1994, 72, 423-434. Kopra, N., Scholz-Ahrens, K.E. and Barth, C A . Effect of casein phosphopeptides on utilization of calcium in vitamin D-replete and vitamin D-deficient rats. Milchwissenschaft 1992, 47, 488-492. Laemmli, U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970, 227, 680-685. Larson, R.A. Naturally Occurring Antioxidants, Lewis Publishers: New York, 1997. Lee, S.K., Han, J.H. and Decker, E.A. Antioxidant activity of phosvitin in phosphatidylcholine liposomes and meat model systems. Journal of Food Science 2002, 57,37-41. Lee, Y.S., Noguchi, T. and Naito, H. Phosphopeptides and soluble calcium in the small intestine of rats given a casein diet. British Journal of Nutrition 1980, 43, 467. 156 Lee, Y.S., Noguchi, T. and Naito, H. Intestinal absorption of calcium in rats given diets containing casein or amino acid mixture, the role of casein phosphopeptides. British Journal of Nutrition 1983, 49, 67-76. Lee, Y.S., Park, G. and Naito, H. Supplemental Effect of Casein Phosphopeptides (CPP) on the Calcium Balance of Growing Rats. Journal of Japanese Society of Nutrition and Food Science 1992, 45, 333-338. Lesse, A.J., Campagnari, A.A. , Bittner, W.E. and Apicella, M.A. Increased resolution of lipopolysaccharides and lipooligosaccharides utilizing tricine-sodium dodecyl sulfate-polyacryl amide gel electrophoresis. Journal of Immunological Methods 1990,126, 109-117. Li , H., Kramer, P.M., Lubet, R.A., Steele, V.E. , Kelloff, G.J. and Pereira, M.A. Effect of calcium on azoxymethane-induced aberrant crypt foci and cell proliferation in the colon of rats. Cancer Letters 1998,124, 39-46. Li , Y., Tome, D. and Desjeux, J.F. Indirect effect of casein phosphopeptides on calcium absorption in rat ileum in vitro. Reproduction, nutrition, developpement 1989, 29, 227-233. Liu, Z., Tomotake, H., Wan, G , Watanabe, H. and Kato, N. Combined effect of dietary calcium and iron on colonic aberrant crypt foci, cell proliferation and apoptosis, and fecal bile acids in 1,2-dimethylhydrazine-treated rats. Oncology Reports 2001, 8, 893-897. Lindmark-Mansson, A.B. Antioxidative factors in Milk. British Journal of Nutrition 2000, 84, S103-S110. Lingnert, H. and Eriksson, C E . Antioxidative Maillard reaction products. II. Products fro sugars and peptides or protein hydrolysates. Journal of Food Processing and Preservation 1980, 4, 173-181. Lonnerdal, B. Milk Proteins and Metabolic Requirements of Trace Elements, Minerals, and Vitamins. In Milk Proteins: Nutritional, Clinical, Functional and Technological Aspects, Barth, C.A., Schlimme, E. , Eds. and Steinkopff Verlag Darmstadt: Germany, 1989. Lonnerdal, B. Dietary Factors Influencing Zinc Absorption. Journal of Nutrition 2000,130, 1378S-1383S. Lonnerdal, B. and Glazier, C An approach to assessing trace element bioavailability from milk in vitro. Extrinsic labeling and proteolytic degradation. Biological Trace Element Research 1989,19, 57-69. Ma, Z.J. and Yamaguchi, M . Synergistic Effect of Genistein and Casein Phosphopeptides on Bone Components in Young and Elderly Rats. Journal of Health Science 2000, 46, 474-479. 157 Mahe, S., Huneau, J.F., Marteau, P., Thuillier, F. and Tome, D. Gastroileal nitrogen and electrolyte movements after bovine ingestion in humans. American Journal of Clinical Nutrition 1992, 56, 410-416. Mahe, S., Benamouzig, R., Gaudichon, C., Huneau, J.-F., De Cruz, I., Rautureau, J. and Tome, D. Nitrogen movements in the upper jejunum lumen in humans fed low amounts of casein and P-lactoglobulin. Gastroenterolgie Clinique et Biologique 1995, 19, 20-26. Mahe, S., Roos, N., Benamouzig, R., Davin, L. , Luengo, C , Gagnon, L. , Gausserges, N., Rautureau, J. and Tome, D. Gastrojejunal kinetics and the digestion of the [15N] P-lactoglobulin and casein in humans: the influence of the nature and quantity of the protein. American Journal of Clinical Nutrition 1996, 63, 546-552. Mancuso, J.R., McClements, D.J. and Decker, E.A. The effects of surfactant type, pH, and chelators on the oxidation of salmon oiLin-water emulsions. Journal of Agricultural and Food Chemistry 1999, 47, 4112-4116. Manson, W. and Annan, W.D. The structure of a phosphopeptide derived from P-Casein. Archives of Biochemistry and Biophysics 1971,145, 16-26. Manson, W. and Cannon, J. The reaction of ccsi- and P-casein with ferrous ions in the presence of oxygen. Journal of Dairy Research 1978, 45, 59-67. Matsui, T., Yano, H., Awano, T., Harumoto, T. and Saito, Y. The Influences of Casein Phosphopeptides on Metabolism of Ectopic Bone Induced by Decalcified Bone Matrix Implantation in Rats. Journal of Nutritional Sciences and Vitaminology 1994, 40, 137-145. Maurer, H.R. Towards serum-free, chemically defined media for mammalian cell culture. In Animal Cell Culture. A Practical Approach., Freshney, R.I., Ed., Oxford University Press: New York, 1992. McDonagh, D. and FitzGerald, R. Production of Caseinophosphopeptides (CPPs) from Sodium Caseinate Using a Range of Commercial Protease Preparations. International Dairy Journal 1998, 8, 39-45. McSherry, C.K., Cohen, B.I., Bokkenheuser, V.D., Mosbach, E.H. , Winter, J., Matoba, N. and Scholes, J. Effects of calcium and bile acid feeding on colon tumors in the rat. Cancer Research 1989, 49, 6039-6043. Mei, L. , Decker, E.A. and McClements, D.J. Evidence of iron association with emulsion droplets and its impact on lipid oxidation. Journal of Agricultural and Food Chemistry 1998, 46, 5072-5077. Mei, L. , McClements, D.J. and Decker, E.A. Lipid oxidation in emulsions as affected by charge status of antioxidants and emulsion droplets. Journal of Agricultural and Food Chemistry 1999, 47, 2267-2273. 158 Meisel, H. and Frister, H. Chemical Characterization of a Caseinophosphopeptide Isolated from in vivo Digests of a Casein Diet. Biological Chemistry Hoppe-Seyler 1988, 369, 1275-1279. Meisel, H. and Frister, H. Chemical characterization of bioactive peptides from in vivo digests of casein. Journal of Dairy Research 1989a, 56, 343-349. Meisel, H. and Frister, H. Isolation and Chemical Characterization of a Phosphopeptide from In Vivo Digests of Casein. In Milk Proteins: Nutritional, Clinical, Functional and Technological Aspects, Barth, C.A., Schlimme, E. , Eds., Steinkopff, Darmstadt and Springer: New York, 1989b. Meisel, H. and Schlimme, E. Milk Proteins: precursors of bioactive peptides. Trends in Food Science and Technology 1990,1, 41-43. Meisel, H. Biochemical properties of bioactive peptides derived from milk proteins: Potential nutraceuticals for food and pharmaceutical applications. Livestock Production Science 1997a, 50, 125-138. Meisel, H. Biochemical Properties of Regulatory Peptides Derived from Milk Proteins. Biopolymers 1997b, 43, 119-128. Meisel, H. Overview on Milk Protein-derived Peptides. International Dairy Journal 1998a, 8, 363-373. Meisel, H. and Olieman, C. Estimation of calcium-binding constants of casein phosphopeptides by capillary zone electrophoresis. Analytica Chimica Acta 1998, 372, 291-297. Mellander, O. The physiological importance of casein phosphopeptide calcium salts. II. Peroral calcium dosage of infants. Acta Society Medicine Uppsala 1950, 55, 247-255. Mellander, O. Phosphopeptides: Chemical Properties and Their Possible Role in the Intestinal Absorption of Metals. In The Transport of Calcium and Strontium Across Biological Membranes, Wasserman, R.H., Ed., Academic Press: New York, 1963. Miller, J.G. and Fry, S.C. Characteristics of xyloglucan after attack by hydroxyl radicals. Carbohydrate Research 2001, 332, 389-403. Morrissey, P.A. and O'Brien, N.M. Dietary Antioxidants in Health and Disease. International Dairy Journal 1998, 8, 463-472. Mosmann, T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. Journal of Immunological Methods 1983, 65, 55-63. 159 Mykkanen, H.M. and Wasserman, R.H. Enhanced Absorption of Calcium by Casein Phosphopeptides in Rachitic and Normal Chicks. Journal of Nutrition 1980,110, 2141-2148. Nagasawa, T., Yuan, Y.V. and Kitts, D.D. Casein Phosphopeptides enhance paracellular calcium absorption but do not alter temporal blood pressure in normotensive rats. Nutrition Research 1991,11, 819-830. Naito, H. , Kawakami, A. and Imamura, T. In vivo Formation of Phosphopeptide with Calcium-binding Property in the Small Intestinal Tract of the Rat Fed on Casein. Agricultural Biology and Chemistry 1972, 36, 409-415. Naito, H. and Suzuki, H. Further Evidence for the Formation in vivo of Phosphopeptide in the Intestinal Lumen from Dietary (3-Casein. Agricultural Biology and Chemistry 1974, 38, 1543-1545. Naito, H., Noguchi, T., Sato, R., Tsuji, K. and Hidaka, H. Patent [90-7616]. 1990. Japan. 90. Nakamura, S., Kato, A. and Kobayashi, K. New antimicrobial characteristics of lysozyme-dextran conjugate. Journal of Agricultural and Food Chemistry 1991, 39, 647-650. Nakamura, S., Kato, A. and Kobayashi, K. Enhanced Antioxidative Effect of Ovalbumin Due to Covalent Binding of Polysaccharides. Journal of Agricultural and Food Chemistry 1992, 40, 2033-2037. Nakamura, S., Kobayashi, K. and Kato, A. Role of positive charge of lysozyme in the excellent emulsifying properties of Maillard-type lysozyme-polysaccharide conjugate. Journal of Agricultural and Food Chemistry 1994, 42, 2688-2691. Nakamura, S., Ogawa, M . , Nakai, S., Kato, A. and Kitts, D.D. Antioxidant Activity of a Maillard-Type Phosvitin-Galatomannan Conjugate with Emulsifying Properties and Heat Stability. Journal of Agricultural and Food Chemistry 1998, 46, 3958-3963. Nakamura, S., Saito, M . , Goto, T., Saeki, H., Ogawa, M . , Gotoh, M . , Gohya, Y. and Hwang, J.K. Rapid Formation of Biologically Active Neoglycoprotein from Lysozyme and Xyloglucan Hydrolysates through Naturally Occurring Maillard Reaction. Journal of Food Sciences and Nutrition 2000, 5, 65-69. Nakamura, S. and Kato, A. Multi-functional biopolymer prepared by covalent attachment of galactomman to egg-white proteins through naturally occurring Maillard reaction. Nahrung 2000, 44, 201-206. Nunez, M.T., Tapia, V., Toyokuni, S. and Okada, S. Iron-induced oxidative damage in colon carcinoma (Caco-2) cells. Free Radical Research 2001, 34, 57-68. Ono, T., Ohotawa, T. and Takagi, Y. Complexes of Casein Phosphopeptide and Calcium Phosphate Prepared from Casein Micelles by Tryptic Digestion. Bioscience, Biotechnology and Biochemistry 1994, 58, 1376-1380. 160 Ono, T., Takagi, Y. and Kunishi, I. Casein phosphopeptides from casein micelles by successive digestion with pepsin and trypsin. Bioscience, Biotechnology and Biochemistry 1998, 62, 16-21. Otani, H., Kihara, Y. and Park, M . The immunoenhancing property of a dietary casein phosphopeptide preparation in mice. Food and Agricultural Immunology 2000,12, 165-173. Pageot, L.P., Perreault, N., Basora, N., Francoeur, C , Magny, P. and Beaulieu, J.F. Human cell models to study small intestinal functions: recapitulation of the crypt-villus axis. Microscopy Research and Technique 2000, 49, 394-406. Park, O. and Allen, J.C. Antigenicity of Casein Phosphopeptides prepared with immobilized glutamic acid-specific endopeptidase or trypsin. Nutrition Research 2000, 20, 359-370. Park, O. and Allen, J.C. Preparation of phosphopeptides derived alpha s-casein and beta-casein using immobilized glutamic acid-specific endopeptidase and characterization of their calcium binding. Journal of Dairy Science 1998, 81, 2858-2865. Pazianas, M . , Adebanjo, O.A., Shankar, V.S., James, S.Y., Colston, K.W., Maxwell, J.D., Zaidi, M . Extracellular cation sensing by the enterocyte: prediction of a novel divalent cation "receptor". Biochemical and Biophysical Research Communications 1995, 210, 948-53. Pellegrini, N., Ying, M . and Rice-Evans, C. Screening of dietary carotenoids and carotenoid-rich fruits extract for antioxidant activities applying 2,2'-azobis(3-ethylbenzothine-6-surfonic acid) radical cation decolorization assay. Methods Enzymology 1999, 299, 389. Perich, J.W., Black, C.L. , Huq, N.L. and Reynolds, E.C. Epitope Analysis of the Multiphosphorylated Peptide asi-Casein (59-79). Journal of Peptide Science 1999, 5, 221-233. Peters, T.J., Apt, L. and Ross, J.F. Effect of phosphates upon iron absorption studied in normal human subjects and in an experimental model using dialysis. Gastroenterology 1971, 61, 315-322. Peres, J.M., Bouhallab, S., Bureau, F., Maubois, J.L., Arhan, P. and Bougie, D.L. Absorption digestive du fer lie au caseinophosphopeptide 1-25 de la (3-caseine. Lait 1997, 77, 433-440. Peres, J.-M., Bouhallab, S., Petit, C , Bureau, F., Maubois, J.L., Arhan, P. and Bougie, D.L. Improvement of zinc intestinal absorption and reduction of zinc/iron interaction using metal bound to the caseinophosphopeptide 1-25 of 6-casein. Special Care in Dentistry 1998, 38, 465-472. 161 Peres, J.M., Bouhallab, S., Bureau, F., Neuville, D., Maubois, J.L., Devroede, G., Arhan, P. and Bougie, D.L. Mechanisms of absorption of caseinophosphopeptide bound iron. Journal of Nutritional Biochemistry 1999a, 10, 215-222. Peres, J.M., Bouhallab, S., Bureau, F., Maubois, J.L., Arhan, P. and Bougie, D.L. Reduction of Iron/Zinc interactions using metal bound to the caseinophosphopeptides 1-25 of |3-casein. Nutrition Research 1999b, 19, 1655-1663. Pointillart, A. and Gueguen, L. Absence of effect of the incorporation of a milk phosphopeptide on the utilization of calcium and phosphorus in the young pig. Reproduction, nutrition, developpement 1989, 29, 477-486. Porter, N.A., Wolf, R.A. and Weenen, H. The free radical oxidation of polyunsaturated lecithins. Lipids 1980,15, 163-167. Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M . and Rice-Evans, C. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biology and Medicine 1999, 26, 1231-1237. Reddy, M.B., Ffurrell, R.F. and Cook, J.D. Estimation of nonheme-iron bioavailability from meal composition. American Journal of Clinical Nutrition 2000, 71, 937-943. Reeves, R.E. and Latour, N.G. Calcium Phosphate Sequestering Phosphopeptide from Casein. Science 1958, 128, 472-472. Reynolds, E.C. , Riley, P.F. and Adamson, N.J. A selective precipitation purification procedure for multiple phosphoseryl-containing peptides and methods for their identification. Analytical Biochemistry 1994, 217, 277-284. Reynolds, E.C. , Cain, C.J., Webber, F.L., Black, C L , Riley, P.F. and Johnson, I.H. Anticariogenicity of tryptic casein and synthetic-phosphopeptides in the rat. Journal of Dental Research 1995, 74, 1272-1279. Reynolds, E.C. Remineralization of Enamel Subsurface Lesions by Casein Phosphopeptide-stabilized Calcium Phosphate Solutions. Journal of Dental Research 1997, 76, 1587-1595. Ribadeau-Dumas, B. Structure and Variability of Milk Proteins. In Milk proteins: nutritional, clinical, functional and technological aspects; Barth,C.A.; Schlimme,E., Eds.; Steinkopff Verlag Darmstadt: Germany, 1989. Richardson, T. and Korycka-Dahl, M . Lipid Oxidation. In Developments in Dairy Chemistry -- 2; Fox, P.F., Ed.; Applied Science Publishers Ltd.: London, 1983. Reynolds, E.C. Anticariogenic complexes of amorphous calcium phosphate stabilized by casein phosphopeptides. Special Care in Dentistry 1998,18, 8-16. 162 Roberts, P.R., Burney, J.D., Black, K.W. and Zaloga, G.P. Effect of chain length on absorption of biologically active peptides from the gastrointestinal tract. Digestion 1999, 60, 332-337. Rongen, H.A.H. , Bult, A. and van Bennekom, W.P. Liposomes and immunoassays. Journal of Immunological Methods 1997, 204, 105-133. Rose, R.K. Effects of an anticariogenic casein phosphopeptide on calcium diffusion in streptococcal model dental plaques. Archives of Oral Biology 2000, 45, 569-575. Rossi, A., Poverini, G., Lullo, D., Modesti, A., Modica, A. and Scarino, M.L. Heavy Metal Toxicity Following Apical and Basolateral Exposure in the Human Intestinal Cell Lin Caco-2. Toxicology in Vitro 1996,10, 27-36. Saito, Y., Yamada, K., Naito, H. and Lee, Y.S. Patent [H05-284939]. 1993. Japan. Saito, Y., Lee, Y.S. and Kimura, S. Minimum Effective Dose of Casein Phosphopeptides (CPP) for Enhancement of Calcium Absorption in Growing Rats. International Journal of Vitaminology Nutritional Research 1998, 68, 335-340. Sandstrom, B., Davidsson, L. , Cederblad, A. and Lonnerdal, B. Oral iron, dietary ligands and zinc absorption. Journal of Nutrition 1985,115, 411-414. Sato, R., Noguchi, T. and Naito, H. The Necessity for the Phosphate Portion of Casein Molecules to Enhance Ca Absorption from the Small Intestine. Agricultural Biology and Chemistry 1983, 47, 2415-2417. Sato, R., Noguchi, T. and Naito, H. Casein Phosphopeptide (CPP) Enhances Calcium Absorption from the Ligated Segment of Rat Small Intestine. Journal of Nutritional Sciences and Vitaminology 1986, 32, 67-76. Sato, R., Shindo, M . , Gunshin, H., Noguchi, T. and Naito, H. Characterization of phosphopeptide derived from bovine |3-casein: an inhibitor to intra-intestinal precipitation of calcium phosphate. Biochimica et Biophysica Acta 1991,1077, 413-415. Sattar Khan, M.A. , Babiker, E.f.E., Azakami, H. and Kato, A. Molecular Mechanism of the Excellent Emulsifying Properties of Phosvitin-Galactomannan Conjugate. Journal of Agricultural and Food Chemistry 1999, 47, 2262-2266. Satue-Gracia, M.T., Frankel, E.N., Rangavajhyala, N. and German, J.B. Lactoferrin in Infant Formulas: Effect on Oxidation. Journal of Agricultural and Food Chemistry 2000, 48, 4984-4990. Sanchez, L. , Ismail, M . , Liew, F.Y. and Brock, J.H. Iron transport across Caco-2 cell monolayers. Effect of transferrin, lactoferrin and nitric oxide. Biochimica et Biophysica Acta 1996,1289, 291-297. 163 Schanbacher, F.L., Talhouk, R.S. and Murray, F.A. Biology and origin of bioactive peptides in milk. Livestock Production Science 1997, 50, 105-123. Schanbacher ,F.L., Talhouk, R.S., Murray, F.A., Gherman, L.I. and Willett, L.B. Milk-Borne Bioactive Peptides. International Dairy Journal 1998, 8, 393-403. Schlimme, E. , Meisel, H. and Frister, H. Bioactive Sequences in Milk Proteins. In Milk Proteins: Nutritional, Clinical, Functional and Technological Aspects, Barth, C.A., Schlimme, E. , Eds., Steinkopff, Darmstadt and Springer: New York, 1989. Schlimme, E. and Meisel, H. Bioactive peptides derived from milk proteins. Structural, physiological and analytical aspects. Die Nahrung 1995, 39, 1-20. Scholz-Ahrens, K. -E. , Kopra, N. and Barth, C.A. Effect of casein phosphopeptides on utilization of calcium in minipigs and vitamin-D-deficient rats. Zeitschrift fur Erndhrungswissenschaft 1990, 29, 295-298. Scholz-Ahrens, K.E. and Schrezenmeir, J. Effects of bioactive substances in milk on mineral and trace element metabolism with special reference to casein phosphopeptides. British Journal of Nutrition 2000, 84, S147-S153. Shah, B.G., Belonje, B. and Paquet, A. The lack of effect of synthetic phosphoseryl peptide on calcium absorption by the rat. Nutrition Research 1990,10, 1331-1336. Shah, N.P. Effects of milk-derived bioactivities: an overview. British Journal of Nutrition 2000, 84, S3-S10. Shahidi, F. Natural Antioxidants: Chemistry, Health Effects, and Applications, AOCS Press: Champaign, IL., 1997. Shepherd, R., Robertson, A. and Ofman, D. Dairy glycoconjugate emulsifier: casein-maltodextrins. Food Hydrocolloids 2000,14, 281-286. Shinmoto, H. , Dosako, S. and Nakajima, I. Anti-oxidant Activity of Bovine Lactoferrin of Iron/Ascorbate Induced Lipid Peroxidation. Bioscience, Biotechnology and Biochemistry 1992, 56, 2079-2080. Shu, Y.W., Sahara, S., Nakamura, S. and Kato, A. Effects of the Length of Polysaccharide Chains on the Functional Properties of the Maillard-Type Lysozyme-Polysaccharide Conjugate. Journal of Agricultural and Food Chemistry 1996, 44, 2544-2548. Slatkavitz, C.A. and Clydesdale, F .M. Solubility of inorganic iron as affected by proteolytic digestion. American Journal of Clinical Nutrition 1988, 47, 487-495. Smith, C.V. and Anderson, R.E. Methods for Determination of Lipid Peroxidation in Biological Samples. Free Radical Biology and Medicine 1987, 3, 344. 164 South, P.K., Lei, X. and Miller, D.D. Meat enhances nonheme iron absorption in pigs. Nutrition Research 2000, 20, 1749-1759. Srigiridhar, K., Nair, K . M . , Subramanian, R. and Singotamu, L. Oral repletion of iron induces free radical mediated alterations in the gastrointestinal tract of rat. Molecular and Cellular Biochemistry 2001, 219, 91-98. Steijins, J. Food mineral fortification with casein phosphopeptide (CPP). Food Ingredients Europe. Conference proceedings 1997. 1998, 69-71. Steijins, J. Milk ingredients as nutraceuticals. International Journal of Dairy Technology 2002,54, 81-88. Suetsuna, K., Ukeda, H. and Ochi, H. Isolation and characterization of free radical scavenging activities peptides derived from casein. Journal of Nutritional Biochemistry 2000,11, 128-131. Swaisgood, H.E. Chemistry of Milk Proteins. In Developments in Dairy Chemistry Volume 1: Proteins, Fox, P.F., Ed., Applied Science Publishers Ltds: Essex, UK, 1982. Swaisgood, H.E. Structural Changes in Milk Proteins. In Milk proteins: nutritional, clinical, functional and technological aspects; Barth, C.A.; Schlimme, E. , Eds.; Steinkopff Verlag Darmstadt: Germany, 1989. Swaisgood, H.E. Review and Update of Casein Chemistry. Journal of Dairy Science 1993, 76, 3054-3061. Tong, L. M . , Sasaki, S. McClements, D.J. and Decker, E.A. Mechanisms of the Antioxidant Activity of a High Molecular Weight Fraction of Whey. Journal of Agricultural and Food Chemistry 2000, 48, 1473-1478. Tsuchita, H., Goto, T., Yonehara, Y. and Kuwata, T. Calcium and Phosphorus Availability from Casein Phosphopeptides in Male Growing Rats. Nutrition Research 1995,15, 1657-1667. Tsuchita, H. , Goto, T., Shimizu, T., Yonehara, Y. and Kuwata, T. Dietary Casein Phosphopeptides Prevent Bone Loss in Aged Ovariectomized Rats. Journal of Nutrition 1996,126, 86-93. Tsuchita, H., Suzuki, H. and Kuwata, T. The effect of casein phosphopeptides on calcium absorption from calcium-fortified milk in growing rats. British Journal of Nutrition 2001, 85, 5-10. Uma Maheswari, S., Ramadoss, C.S., and Krishnaswamy, P.R. Inhibition of Fe(II) catalyzed linoleic acid oxidation and DNA damage by phosvitin. Molecular and Cellular Biochemistry 1997,177, 47-51. 165 Vinas-Salas, J., Biendicho-Palau, P., Pinol-Felis, C , Miguelsanz-Garcia, S. and Perez-Holanda, S. Calcium inhibits colon carcinogenesis in an experimental model in the rat. European Journal of Cancer 1998, 34, 1941-1945. Wakabayashi, H. , Matsumoto, H., Hashimoto, K., Teraguchi, S., Takase, M . and Hayasawa, H. Inhibition of Iron/Ascorbate-Induced Lipid Peroxidation by an N-Terminal Peptide of Bovine Lactoferrin and Its Acylated Derivatives. Bioscience, Biotechnology and Biochemistry 1999,55,955-957. Wang, T.G. , Gotoh, Y., Ho Jennings, M . , Rhoads, C.A. and Aw, T.Y. Lipid hydroperoxide-induced apoptosis in human colonic Caco-2 cells is associated with an early loss of cellular redox balance. FASEB Journal 2000,14, 1567-1576. Wessling-Resnick, M . Iron Transport. Annual Reviews in Nutrition 2000, 20, 129-151. West, D.W. Structure and function of the phosphorylated residues of casein. Journal of Dairy Research 1986, 53, 333-352. Wien, E .M. , Glahn, R.P. and Van Campen, D.R. Ferrous iron uptake by rat duodenal brush border membrane vesicles: Effect of dietary iron level and competing minerals (Zn + 2, M n + 2 , and Ca + 2). Journal of Nutritional Biochemistry 1994, 5, 571-577. Wilson, A.P. Cytotoxicity and viability assays. In Animal Cell Culture, Freshney, R.I., Ed., Oxford University Press: New York, 1992. Yeung, A.C. , Glahn, R.P. and Miller, D.D. Dephosphorylation of sodium caseinate, enzymatically hydrolyzed casein and casein phosphopeptides by intestinal alkaline phosphatase: implications for iron availability. Journal of Nutritional Biochemistry 2001,12, 292-299. Yuan, Y.V. and Kitts, D.D. Confirmation of Calcium Absorption and Femoral Utilization in Spontaneously Hypertensive rats fed Casein Phosphopeptide Supplemented Diets. Nutrition Research 1991,11, 1257-1272. Yuan, Y.V. and Kitts, D.D. Estimation of dietary calcium utilization in rats using a biochemical functional test. Food Chemistry 1992, 44, 1-7. Yuan, Y.V. and Kitts, D.D. Calcium absorption and bone utilization in spontaneously hypertensive rats fed on native and heat-damaged casein and soybean protein. British Journal of Nutrition 1994, 71, 583-603. Yvon, M . and Pelissier, J.P. Characterization and Kinetics of Evacuation of Peptides Resulting from Casein Hydrolysis in the Stomach of the Calf. Journal of Agricultural and Food Chemistry 1987, 35, 148-156. 166 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090381/manifest

Comment

Related Items