Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The provision of passive immunity to colostrum-deprived piglets by bovine or porcine serum immunoglobulins,… Drew, Murray D. 1989

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

Item Metadata

Download

Media
831-UBC_1989_A1 D73.pdf [ 8.55MB ]
Metadata
JSON: 831-1.0098280.json
JSON-LD: 831-1.0098280-ld.json
RDF/XML (Pretty): 831-1.0098280-rdf.xml
RDF/JSON: 831-1.0098280-rdf.json
Turtle: 831-1.0098280-turtle.txt
N-Triples: 831-1.0098280-rdf-ntriples.txt
Original Record: 831-1.0098280-source.json
Full Text
831-1.0098280-fulltext.txt
Citation
831-1.0098280.ris

Full Text

THE PROVISION OF PASSIVE IMMUNITY TO COLOSTRUM-DEPRIVED PIGLETS BY BOVINE OR PORCINE SERUM IMMUNOGLOBULINS, IRON CHELATORS AND VIABLE LEUKOCYTES By Murray D. Drew B.Sc. (Agr), The University of Guelph, 1983 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF ANIMAL SCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1989 © Murray D. Drew, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) i i ABSTRACT This thesis examines the effect of supplementing sow milk replacers fed to colostrum deprived piglets with 1) bovine or porcine immunoglobulins; 2) synthetic iron chelators and 3) viable leukocytes. Piglets require dietary immunoglobulins during the f i r s t day after birth to provide passive systemic immunity. Piglets that did not receive immunoglobulins during this period had survival rates of 19% and 0% in two different exper-iments. Bovine immunoglobulins on day 1 after birth were poorly absorbed from the diet resulting in inadequate plasma immunoglobulin concentrations, low survival and low weight gains compared to piglets that received porcine immuno-globulins on day 1. During the days 2-14, however, piglets receiving either bovine or porcine immunoglobulins were equal in survival and growth rates. The milk protein lactoferrin protects piglets from enteric infections by binding ionic iron making, i t unavailable to bacteria and preventing bacterial growth. Two synthetic iron chelators, ethylenediamine-di-orthohydroxyphenyl acetic acid (EDDA) and N,N'-bis(o-hydroxybenzyl)-ethylenediamine diacetic acid (HBED) have a f f i n i t i e s for iron similar to lactoferrin's and are potential substitutes for lactoferrin in sow milk replacers. The antibacterial properties of lactoferrin, EDDA and HBED were compared in vitro. Lactoferrin and EDDA inhibited the growth of E^ c o l i 0 157 K88 over a 12 hour period while HBED had no effect on the growth of this organism. When EDDA or HBED were fed to piglets on days 2-14, those that received HBED had low survival and growth rates. Piglets that received EDDA had growth rates similar to those receiving porcine immunoglobulins from days 2-14. When EDDA was fed during days 1-14 however piglet growth rates were depressed. This was probably due to higher absorption of EDDA during day 1. Dietary EDDA also increased the excretion of iron in the urine and feces and decreased the incorporation of iron into hemoglobin. The feeding of viable leukocytes derived from abattoir blood to a r t i f i c i a l l y reared piglets fed porcine immunoglobulins resulted in increased c e l l mediated immune responses in 2 of 4 l i t t e r s . The major histocompatibility complex types of the donor and recipient of the leukocytes may be responsible for the inconsistent results. Porcine immunoglobulins (25 mg mir 1 on day 1, followed by either bovine or porcine immunoglobulins (5 mg mir 1) on days 2-14 provided adequate passive immunity, survival and growth rates in colostrum deprived piglets. iv TABLE OF CONTENTS Page ABSTRACT i i L I S T OF TABLES v i L I S T OF FIGURES X ACKNOWLEDGEMENTS x i i INTRODUCTION 1 A REVIEW OF THE IMMUNE SYSTEM 2 The Major Histocompatability Complex 4 Neutrophils and Macrophages 6 T Lymphocytes 8 B Lymphocytes and Immunoglobulins 12 Complement 20 The Ontogeny of the Piglet's Immune System 23 Host Defences in the Neonatal Piglet 25 PROTECTIVE FACTORS I N SOW'S COLOSTRUM AND MILK 28 Viable Cells 30 Immunoglobulins 33 Immunoglobulin Fortified Milk Replacers 41 Lactoferrin 43 Synthetic Iron Chelators 52 Vitamin B12 Binding Protein 55 Lysozyme 56 The Lactoperoxidase System 57 Glycoconjugate and Oligosaccharide Receptor Analogues 58 EXPERIMENT 1 60 Introduction 60 Materials and Methods 60 Results 65 Discussion 67 EXPERIMENT 2 79 Introduction 79 Materials and Methods 79 Results 83 Discussion 84 EXPERIMENT 3 90 Introduction 90 Materials and Methods 90 Results 94 V Discussion • • • 96 EXPERIMENT 4 109 Introduction 109 Materials and Methods 109 Results 112 Discussion 115 EXPERIMENT 5 131 Introduction 131 Materials and Methods 131 Results 134 Discussion 135 EXPERIMENT 6 144 Introduction 144 Materials and Methods 144 Results 148 Discussion 149 GENERAL DISCUSSION 161 CONCLUSIONS 164 LITERATURE CITED 165 v i LIST OF TABLES Page Table 1. Classification of placentation according to the intervening tissues and correlation with time of transfer of immunity from mother to offspring 1 Table 2. Nonspecific host defense factors 3 Table 3. Ontogeny of the immune system of the fetal piglet 24 Table 4. Approximate chemical analysis of sow's colostrum and milk ... 28 Table 5 . Differential c e l l counts for the mammary secretions of sows in which bacterial infection was absent 32 Table 6. Immunoglobulin levels in serum, colostrum, milk and intestinal juice of pigs 34 Table 7 . Concentration of lactoferrin in various secretions of the bovine mammary gland 50 Table 8. Comparison of the st a b i l i t y constants of HBED and EDDA chelates 55 Table 9. Experimental protocol for studying the administration of immunoglobulins to piglets ( Experiment 1) 61 Table 10. Composition of freeze dried bovine and porcine immunoglobulin concentrates 72 Table 11. The effect of bovine or porcine immunoglobulins on the survival of colostrum deprived piglets (Experiment 1) 73 Table 12. The effect of bovine or porcine immunoglobulins on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 1) 73 Table 13. The effect of bovine or porcine immunoglobulins on the average daily gains of colostrum deprived piglets (Experiment 1) 74 Table 14. The effect of bovine or porcine immunoglobulins on the body weights of colostrum deprived piglets (Experiment 1) 74 Table 15. The effect of bovine or porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets levels (Experiment 1) 75 v i i Table 16. Experimental protocol for testing the effect of lactoferrin, EDDA and HBED on the growth rate of E^ . c o l i 0 157 K88 (Experiment 2) 80 Table 17. The effect of lactoferrin, EDDA and HBED with and without porcine immunoglobulins, on the mean Log(CFU) mir 1 of E^ c o l i 0 157 K88 (Experiment 2) 88 Table 18. Experimental protocol for testing the effect of EDDA and HBED on the performance of a r t i f i c i a l l y reared piglets (Experiment 3) 91 Table 19. The effect of EDDA or HBED, with or without porcine immunoglobulins on the survival of colostrum deprived piglets (Experiment 3) 102 Table 20. The effect of EDDA or HBED, with or without porcine immunoglobulins on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 3) 103 Table 21. The effect of EDDA or HBED, with or without porcine immunoglobulins on the average daily gains of colostrum deprived piglets (Experiment 3) 104 Table 22. The effect of EDDA or HBED, with or without porcine immunoglobulins on the body weights of colostrum deprived piglets (Experiment 3) 105 Table 23. The effect of EDDA or HBED, with or without porcine immunoglobulins on the plasma iron and total iron binding capacity (TIBC) of colostrum deprived piglets (Experiment 3) 106 Table 24. The effect of EDDA or HBED, with or without porcine immunoglobulins on the packed c e l l volume and hemoglobin concentrations of colostrum deprived piglets (Experiment 3) 107 Table 25. The effect of EDDA or HBED, with or without porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 3) 108 Table 26. Experimental protocol for studying the administration of EDDA with porcine or bovine immunoglobulins (Experiment 4) 111 Table 27. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the survival of colostrum deprived piglets (Experiment 4) 123 v i i i Table 28. The effect of EDDA , with and without bovine or porcine immunoglobulins, on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 4) 124 Table 29. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the average daily gains of colostrum deprived piglets (Experiment 4) 125 Table 30. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the body weights of colostrum deprived piglets (Experiment 4) 126 Table 31. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma iron and total iron binding capacity (TIBC) of colostrum deprived piglets (Experiment 4) 127 Table 32. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the packed c e l l volume and hemoglobin concentrations of colostrum deprived piglets (Experiment 4) 128 Table 33. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 4) 129 Table 34. The effect of EDDA on individual piglets on the day of " F e injection (Experiment 5) 139 Table 35. The effect of EDDA on excretion and distribution of 3 9 F e (Experiment 5) 140 Table 36. Experimental protocol used to study the effect of sow or a r t i f i c i a l rearing on colostrum deprived piglets (Experiment 6) 145 Table 37. The effect of sow or a r t i f i c i a l rearing on piglet survival (Experiment 6) 156 Table 38. The effect of sow or a r t i f i c i a l rearing on piglet average daily gains (Experiment 6) 157 Table 39. The effect of sow or a r t i f i c i a l rearing on mean piglet weights (Experiment 6) 157 Table 40. The effect of sow or a r t i f i c i a l rearing on plasma PIgG levels (Experiment 6) 158 ix Table 41. The effect of sow or a r t i f i c i a l rearing on piglet intradermal response to PHA at 3 weeks of age (Experiment 6) 159 Table 42. The effect of sow or a r t i f i c i a l rearing on piglet intradermal response to PHA at 24 hours for treatment x block interaction (Experiment 6) 160 Table 43. The average daily gains and survival of piglets in Experiments 1, 3, 4 and 6 164 X LIST OF FIGURES Page Figure 1- Hemopoietic stem c e l l differentiation 5 Figure 2. Structure of an IgG molecule 13 Figure 3. The basic structure of the immunoglobulin classes 15 Figure 4. The ontogeny of B cells 17 Figure 5. The genetic basis of antibody production 18 Figure 6 . The classical, alternate and terminal pathways of the complement cascade 21 Figure 7 . The concentration of IgG, IgM and IgA in the sera of naturally reared piglets 36 Figure 8 . The covalent structure of human monomeric, dimeric and secretory IgA 38 Figure 9 . The preparation of bovine lactoferrin from milk 44 Figure 10. The molecular structure of EDDA and HBED 53 Figure 11. Gel electrophoresis of polyphosphate fractions from porcine serum 76 Figure 12. Gel electrophoresis of polyphosphate fractions from bovine serum 77 Figure 13. The effect of bovine or porcine immunoglobulins on the plasma immunoglobulin concentration of colostrum deprived piglets (Experiment 1) 78 Figure 14. The effect of lactoferrin, EDDA and HBED with and without porcine immunoglobulins, on the mean Log (CPU) mL-1 of J L c o l i 0 157 K88 (Experiment 2) 89 Figure 15. The effect of EDDA or HBED, with or without porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 3) 109 Figure 16. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 4) 130 xi Figure 17. The effect of EDDA on plasma iron disappearance (Experiment 5) 141 Figure 18. The effect of EDDA on the incorporation of " F e into red blood cells (Experiment 5) 142 Figure 19. The effect of EDDA on urinary and fecal excretion of 3 9Fe (Experiment 5) 143 Figure 20 . The effect of sow or a r t i f i c i a l rearing on piglet plasma IgG (Experiment 6) 161 x i i ACKNOWLEDGEMENTS I would like to thank Dr. Bruce Owen for his ideas, support and kindness over the last four years. The members of my committee, Dr. R. Blair, Dr. C. R. Krishnamurti, Dr. J.A. Shelford and Dr. B.J. Skura have a l l given me excellent advice and encouragement throughout study. Many sleepless nights were spent feeding piglets by Richard Whiting, Margaret Crowley, Shane Gumprich and Tri c i a Arnold. I thank them for their dedication. Gilles Galzy and Frances Newsome were always there with help and good ideas when problems arose. In particular, I would like to thank Irene Bevandick for her commitment and excellent technical assistance. Financial assistance provided by the Productivity Enhancement Program of the Canada/British Columbia Agri-Food Regional Development Subsidiary Agreement and by the Natural Science and Engineering Research Council i s gratefully ac-knowledged. Finally, the love and support of my wife Bev have made the last four years possible and enjoyable. 1 INTRODUCTION The newborn piglet i s a physiologically mature and capable creature compared to the young of many other species. It can walk, see and feed i t s e l f soon after birth. A l l of the elements of the adult immune system are present and the piglet is able to mount an immune response, but only a primary immune response (Salmon 1987). The primary response of the immune system on f i r s t exposure to a pathogen is slow to develop and provides inadequate protection against infection. In addition to inadequate active immunity, the epitheliochorial placentation of the pig does not allow placental transfer of immunoglobulins (see Table 1). The piglet is thus born with no passive protection and is completely dependent on sow's colostrum and milk for i t s survival. Table 1. C l a s s i f i c a t i o n of p l a c e n t a t i o n according to the i n t e r v e n i n g t i s s u e s and the c o r r e l a t i o n with t i a e of t r a n s f e r of immunity f r o a mother to o f f s p r i n g ( S t e r z l and S i l v e r s t e i n 19671 L. Uterine t i s s u e s F e t a l t i s s u e s Tiae of Transmission Type of P l a c e n t a t i o n A n i a a l Endo- Connec- Ep i t h e - Tropho- Connec- Endothe t h e l i u a t i v e l i u a b l a s t t i v e l i u a P r enatal P o s t n a t a l E p i t h e l i o c h o r i a l Horse, Pig + + + Syndesaochorial Sheep, Cow + + 0 End o t h e l i o c h o r i a l Cat, Dog + 0 0 Heaoendothelial Rat, House 0 0 0 Heaoendothelial Rabbit 0 0 0 Heaochorial Man, Monkey 0 0 0 0 0 + + +++ +++ +++ (36 hrs) +++ (36 hrs) ++ (10 Days) ++ (20 Days) 0 0 1 Symbols employed i n t h i s table are: +++ = s o l e c o n t r i b u t i o n . 0 = absence; + = pres nee; ++ = major c o n t r i b u t i o n ; 2 Fortunately for the piglet, sow colostrum is usually available to provide this needed protection. However, the death of the sow, mastitis, agalactia, and large l i t t e r s can mean that piglets do not receive adequate colostral passive immunity. A method of replacing sow's milk and colostrum would be a valuable tool to the swine producer. To develop such a system requires understanding the development of the piglet's immune system and i t s regulation by the sow's milk and colostrum. A REVIEW OF THE IMMUNE SYSTEM The Webster New American Dictionary (1972) defines the word immune as "exempt, as from a disease". The words immunity and immune system immediately conjure up the idea of a large host destroying a small parasite (bacteria, virus etc.) that has invaded i t s body. In a larger sense, immunity also affects the host's reaction to food proteins, insect stings and tissue transplants. The immune system consists of a l l of the physical structures and physiological responses that affect interactions between the host and the rest of the uni-verse. Non-specific aspects of immunity are usually overlooked in discussions of the immune system (see Table 2). Non-specific host defense is at least as important as specific immunity, however. Intact body surfaces, normal bacterial flora etc. reduce the number of interactions the specific immune system has with the bacteria and viruses. In addition, i t i s d i f f i c u l t to t e l l where non-specific immunity ends and specific responses begin as non-specific defenses interact with specific ones. The specific immune system is responsible for distinguishing self from non-self and mounting an appropriate response. The response may be tolerance for normal body flora or food proteins. It may be destructive for pathogenic bacteria or virus infected c e l l s . This is a complex task, so complex that the immune system i s comparable to the human brain in i t s intricacy. The immune system is like the brain in many respects. It can process information, i t has memory and is able to respond to antigens i t has never encountered before. Table 2. Nonspecific host defense factors (Beisel, 1984). Host system Passive Factors Actively responsive Factors  Microbiological Physical Secretions (internal and external) Physiological Humoral Normal body flora Anatomical surface barriers Anatomical pathways Cili a r y cleansing motion Gastric and urinary acidity Mucins and their enzymes Fatty acids in secretions Lysozyme in secretions Lactoferrin Salivary enzymes H2O2 Surfactants Age related influences Sex related influences Genetic effects Circadian rhythms Tissue pH Plasma transport proteins Plasma redox potential Divalent cations: Ca 2 + and Mg2+ Cough reflex Vomiting reflex Intestinal peristalsis Interferons Tuftsin Circulating S-lysins Mast c e l l products e.g., Histamine Heparin, Serotonin, Anaphylactic factors, Enzymes Lymphokines and Thymic hormones, Biologically active small molecules Inflammatory reactions Cellular phagocytic and bactericidal activity Fever generation Metabolic responses eg., amino acids, carbohydrates,lipids, electrolytesminerals, Coagulation system Complement system Kinin system Plasma acute-phase reactant proteins  4 Antigens are the basic unit to which the immune system responds. To i n i t i a t e an immune response an antigen must have several properties. They are normally large complex molecules with stable 3-dimensional structures (Tizard 1982). This makes most protein molecules ideal antigens. Carbohydrates and lipids are usually too small and flexible to make good antigens. The immune system consists of several different types of cells and some specialized organs, for example the thymus. The cells are a l l derived from pluripotent stem cells found in the bone marrow. These stem cells have several important properties. They are self-renewing and are capable of differentiation into a l l c e l l groups derived from the bone marrow. Because the stem cells are continually renewing themselves, they are extremely sensitive to radiation. Their destruction i s the primary cause of death from exposure to radiation. The stem cells differentiate into two lineages of cells (see Figure 1). The cells of the myeloid lineage include erythrocytes, platelets, granulocytes and macrophages. The lymphoid lineage includes T-lymphocytes and B-lymphocytes. The macrophages, T cells and B cells are the cellular components of the immune system. The Major Histocompatibility Complex The major histocompatibility complex (MHC) is a group of genes that encode the MHC antigens. The MHC in swine is termed the Swine Lymphocyte Antigen (SLA). It occurs on chromosome 7 (Vaiman 1987) and consists of three different classes of genes. Class I SLA genes are the A, B and C l o c i (Vaiman et a l . 1986). These genes code for the class I SLA antigens. These antigens are glycoproteins with a molecular weight of about 45 Kd. A l l nucleated cells in the body express these 5 Figure 1. Hemopoietic stem c e l l differentiation (based on Cooper et a l . 1984). Pluripotent Stem Cell Myeloid Lineage / \ Lymphoid Lineage Myeloid Stem Cell Lymphoid Stem Cell / 1 Erythrocyte Platelets Granulocyte Progenitor Lymphocytes B Lymphocytes / \ Neutrophils Macrophages Plasma Cell antigens. The class I antigens are released into the cytosol of the c e l l where they bind with peptides present there (Maryanski et a l . 1986). These peptides may be normal c e l l constituents or abnormal products of viruses or cancer c e l l s . The complex of class I molecule and peptides then migrate to the surface of the c e l l . Cytotoxic T cells recognize foreign polypeptides and lyse the c e l l . Cytotoxic T cells also lyse cells that display foreign class I MHC molecules. This is the mechanism that causes the rejection of transplants between non-compatible donors. It is also responsible for graft-versus-host disease. Graft-versus-host disease occurs when immunocompetent cells transferred to a MHC class I incompatible host react against the host tissues. Normally the immune system of the host can resist this attack. In immunodeficient hosts such as neonatal 6 piglets ,however, the graft can cause considerable damage to host tissues. More than 30 different class I SLA antigens have been characterized in swine (Vaiman 1987) with approximately 10-12 alleles per locus. This means that over 1000 different haplotypes are possible. Only 62 different haplotypes have been found however. This is due to high linkage disequilibrium between the class I genes. The class I haplotype of pigs affects a number of production traits including ovulation rate (Rothschild et a l . 1984), embryonic survival (Vaiman et a l . 1986) and birth and weaning weights (Rothschild et a l . 1986). There are two class II l o c i in the SLA: the DQ and the DR. Class II antigens occur on lymphocytes and macrophages. They control interactions among these cellular components of the immune system. Macrophages phagocytize foreign particles in the body and digest them. These fragments are bound by class II antigens and the complex migrates to the surface of the macrophage (Babbitt et a l . 1985). The class II haplotype of pigs controls their a b i l i t y to respond to different antigens and susceptibility to disease (Vaiman et a l . 1986). The class III SLA genes code for several components of the complement cascade. This class of SLA genes influences hemolytic complement activity and responsiveness to antigens (Vaiman 1987). Neutrophils and Macrophages Neutrophils are the f i r s t line of defence during an infection. They occur in large numbers in the blood stream and are rapidly mobilized to a site of tissue injury by various chemotactic stimuli (Klebanoff and Clark 1978) . Here they rapidly phagocytize foreign material. Once the particle i s ingested i t is k i l l e d by oxidative metabolites and digested. Neutrophils have limited energy however and cannot maintain their attack for very long. Macrophages are required to sustain the response to infection. 7 Macrophages have 3 major functions: (i) they are important in the elimination of foreign material via phagocytosis; (ii) they present antigen to lymphocytes and ( i i i ) they secrete important bioactive molecules that are important in host defense (Kende 1982; Shevach 1984). Macrophages consist of several different types of c e l l s . After differen-tiation in the bone marrow, immature macrophages enter the blood stream where they circulate (Kende 1982). These cells migrate to the tissues where they become mature macrophages. Some types of macrophages become resident in body tissues. These include Kupffer's cells in the liver and alveolar cells in the lung. These cells are stationed at strategic points of the body to remove foreign material and microorganisms from the blood stream. Macrophages migrate to sites of infection in response to bacterial products, complement cascade products and factors released by dying neutrophils (Tizard 1986a). Once at the site, macrophages bind to foreign bacteria, ingest them and k i l l them. The production of hydroxyl radicals, singlet oxygen and hypochloride ion during the respiratory burst k i l l s the ingested microbe. As the macrophages mature they develop membrane receptors that aid in their physiological functions (Silberberg-Sinakin et a l . 1980). For phagocytosis of bacteria to take place there must be adherence between the macrophage and the bacteria. Bacterial slime coats, capsules and carbohydrates decrease adherence so phagocytosis is less efficient. Immunoglobulins and the C3b complement fragment can bind to these bacterial surfaces. The receptors for the Fc portion of IgG and IgM and for the C3b complement fragment increase the adherence between macrophages and bacteria. Factors that increase adherence between phagocytic cells and bacteria are called opsonins. Macrophages also have very high levels of membrane Class II MHC antigens. 8 This relates to their role in presenting antigens to T c e l l s . After foreign antigens are ingested, they are denatured and cleaved in lysosomes (Babbitt et a l . 1985). These fragments are then bound by class II MHC molecules and the complex migrates to the c e l l membrane. T cells specific for this complex bind to i t . This is the f i r s t step in the activation of the T helper c e l l . T cells are incapable of responding to many antigens unless macrophages f i r s t process and present the antigen (Waldron et a l . 1973). Only macrophages that express Class II MHC on their c e l l surfaces can present antigen to T cells (Yamashita and Schevach 1977). The more Class II antigen present on the c e l l surface, the more efficient the macrophage is in stimulating an immune response. The macrophages of neonates have low levels of Class II antigens (Tizard 1986a) and this i s partially responsible for their immunodeficient condition. The presentation of antigen to a T c e l l by a macrophage results in the release of Interleukin-1 by the macrophage. This stimulates T c e l l activity and also stimulates B cells (Kende 1982). Macrophages secrete a wide variety of different products which include antibacterial proteins like lysozyme and lactoferrin, complement components and immunoregulatory factors like inter-leukin-1. The functions of the macrophage include roles in non-specific host defense and in the immune system. It coordinates the activity of both systems so they operate ef f i c i e n t l y in the protection of the host. T Lymphocytes The lymphoid lineage of the pluripotent stem cells consists of 2 populations of lymphocytes, the B lymphocytes and the T lymphocytes. The two populations arise from the same group of cells and are indistinguishable in appearance. Both have receptors for specific antigens and the DNA of these receptors is 9 rearranged in somatic c e l l s . Here the similarities end. The functions of the two groups of lymphocytes are completely different. B cells produce soluble antibodies which bind to antigen. T cells regulate the functioning of the immune system and are responsible for destroying virus infected cells and cancer c e l l s . T cells can be divided into subpopulations based upon the expression of several markers. The terminology of these markers is different for different species. Most studies of T cells use mice so the terminology for murine T cells is used here with appropriate references to terminology for pigs. A l l mature murine T cells express the CD3 complex (Samuelson et a l . 1985). This complex consists of 6 or more proteins associated with the T c e l l antigen receptor. Other important markers of T cells include the CD4 and CD8 antigens. These are functional markers (von Boehmer 1988). T cells expressing the CD4 antigen are T helper c e l l s . These CD4+ cells activate other components of the immune system including B cells and cytotoxic T ce l l s . The CD8+ T cells include these cytotoxic T ce l l s . In addition to these functional differences, the CD4+ and CD8+ T cells also differ in the class of MHC molecules they recognize. The CD4+ cells recognize class II MHC molecules while the CD8+ cells recognize class I molecules. Markers for these populations in pigs have been characterized (Salmon 1987) . The CD4 antigen in mice corresponds to the PT4 antigen in pigs and the CD8 antigen corresponds to the PT8 antigen. T cells bearing the PT4 antigen are helper cells while the PT8 T cells are suppressor or effector T Cells. In addition several markers that differentiate porcine T cells from B cells exist. Porcine T cells w i l l form rosettes with sheep red blood c e l l s . This effect is due to a receptor specific to porcine T ce l l s . A monoclonal antibody (MSA4) to this receptor has been synthesized (Pescowitz et a l . 1985). 10 The antigen receptor of T cells differs from immunoglobulin molecules that serve as B c e l l antigen receptors. T c e l l receptors are not secreted by T cells and they do not bind soluble antigens (von Boehmer 1988). This has made characterizing them much more d i f f i c u l t than immunoglobulins. The T c e l l receptor does not recognize native antigen but i t recognizes processed antigen associated with class I or class II MCH molecules. The T c e l l receptor is a heterodimer containing an a and a S chain which are joined by disulfide bridges (Yague et a l . 1985). Both proteins have mole-cular weights of about 40 Kd. The diversity of T c e l l receptor antigen speci-f i c i t i e s is generated by the rearrangement of the genes that code for the a and fi chains (Snodgrass et a l . 1985). The same genes code for the antigen receptors of both CD4+ and CD8+ cells (Dembic et a l . 1986) . This is surprising because the two populations of cells recognize different MHC molecules. The answer to this is that the recognition of antigen by T c e l l requires a whole complex of proteins of which the T c e l l receptor is one part (von Boehmer 1986; Emmrich et a l . 1986). The CD3 complex and the T c e l l receptor are part of the antigen receptor for both CD4+ .and CD8+ T ce l l s . The CD4 molecule is thought to cross link with the a and fi chains of the T c e l l receptor on Class II MHC molecules. The same is true for the CD8+ T cells except that here the CDS protein cross links with the a and fi chains on a class I MHC molecule. The specificity for the antigen bound by the MHC molecule i s provided by the a and fi chains. The specificity for either class I or II molecules is due to the CD4 or CD8 molecules. T lymphocytes are produced in the bone marrow and migrate to the thymus. These immature T cells are CD3" • CD4~ • CD8- and the genes that code for the T c e l l receptor have not been rearranged (von Boehmer 1988). The f i r s t event is 11 the rearrangement of the genes for the a and 2 chains (von Boehmer 1988). After this the cells become CD3+,CD4+,CD8+ or CD3~ • CD4+,CD8+ . These cells then undergo thymic processing in which a l l CD3- cells are destroyed. In addition CD3+ cells that react with self antigens are also destroyed. Less than 1% of the T cells that enter the thymus leave i t (Scollay et a l . 1980). The cells that leave the thymus are CD3+,CD4+,CD8+. These cells are unspecialized. They can be transformed into either CD3+,CD4+,CD8_ or CD3+ ,CD4- • CD8+ cells as required (Shen et a l . 1980). The CD3+,CD4+,CD8~ cells are T helper c e l l s . As mentioned previously, macrophages process and present antigen in the context of class II MHC molecule. When a T helper c e l l that is specific for this presented antigen binds to the macrophage interleukin-1 is released. Interleukin-1 activates the T c e l l and i t begins to divide. The clone of T cells in turn activates B ce l l s . B c e l l activation may be antigen specific or non-specific (Tada 1984) . Antigen specific help involves the binding of the T c e l l to determinants on an antigen bound to the B cells antigen receptor. These T cells also recognize class II MHC molecules on the surface of the B c e l l . The recognition of both specific antigen and the class II molecule causes the T c e l l to release interleukin-2. Inter-leukin 2 causes the T cells to divide and to secrete lymphokines that activate B cells (Miedema and Melief 1985) . T cells can also help B cells by the release of soluble factors. These factors are not antigen specific. They non-selectively stimulate B cells to produce immunoglobulins. Cytotoxic T cells destroy virus infected and cancerous target c e l l s . These T cells are CD3+,CD4"-CD8+ and are MHC class I restricted. Cytotoxic T cells recognize non-self peptides bound to the class I molecules. This causes the T c e l l to secrete perforins (Herberman and Ortaldo 1981). Perforins aggregate 12 together and form hollow tubes. These tubes insert into the membrane of the target c e l l and cause l y s i s . Suppressor T cells are another population of CD3+,CD4_> CD8+ cells (von Boehmer 1988). These cells suppress immune responses by inhibiting the pro-duction of interleukin-2 by T helper cells (Asherson et a l . 1986). B Lymphocytes and Immunoglobulins The second population of lymphocytes is the B c e l l s . B cells receive their name from the fact that they mature in the bursa of Fabricius in birds. For many years i t was unknown where B cells matured in mammals as there was no obvious bursal equivalent. In recent years i t has become apparent that B cells are processed in the bone marrow (Cooper 1981). B cells synthesize and secrete immunoglobulins. Immune responses mediated by immunoglobulins are termed humoral immune responses. Immunoglobulins are glycoproteins consisting of 4 protein chains covalently linked together (Figure 2). There are 2 heavy chains and 2 light chains in each immunoglobulin molecule. The primary structure of both chains shows internal homology (Silverton et a l . 1977). They are made up of blocks of about 110 amino acids called domains (Silverton et a l . 1977). The light chains consist of 2 domains while the heavy chains consist of 4 or 5 domains depending on the class of immunoglobulin. The domains can be subdivided into variable and constant domains. The variable domains are the regions of the immunoglobulin molecule that actually bind to antigen. The constant domains determine the class of the immunoglobulin. Digestive enzymes like papain, w i l l cleave the immunoglobulin molecule Figure 2. Structure of an IgG molecule (Based on Cooper et a l . 1984). VL A N D V I : V A R I A B L E D O M A I N S Ci A N D C i : C O N S T A N T D O M A I N S 14 at the hinge region creating 3 fragments (Porter 1959). The two Fab fragments contain the variable regions of both the heavy and light chains (see Figure 2). They retain their ability to bind antigen but lose the abil i t y to bind complement or bind to receptors on macrophages. The Fc portion of the immuno-globulin molecule w i l l not bind antigen but will bind complement and bind to receptors on macrophages (Fc receptors). The secondary and tertiary structures of each domain are similar. Each domain is made up of two fi-pleated sheet structures connected by an intradomain disulfide bond (Silverton et a l . 1977). The cysteine residues that form the disulfide bond are conserved in a l l immunoglobulin classes. In pigs, there are two types of light chains, lambda and kappa. There are 3 types of heavy chains u, x and a. The heavy chain type determines the class of the immunoglobulin molecule. Immunoglobulins containing u chains are called IgM, x chains give IgG and a chains IgA. In many other species there are also t and 5 heavy chains giving IgE and IgD respectively. . IgM is the immunoglobulin synthesized f i r s t during B c e l l ontogeny (Kincaid 1981). It is synthesized and secreted during a l l primary immune responses. Neonatal piglets do not normally encounter any antigens in utero. Their immune responses are therefore primary immune responses and almost entirely IgM (Allen and Porter 1977). IgM occurs in two forms. IgM found bound to the membranes of B cells is a monomer while serum IgM is a pentamer (see Figure 3) (Cooper et a l . 1984) with the five IgM units arranged radially with the antigen binding sites pointing outwards. The five units of pentameric IgM are joined by protein called the J chain. IgM makes up about 5-10% of serum immunoglobulins. IgM is the most efficient of the immunoglobulin classes at in i t i a t i n g the complement cascade (Borsos and Rapp 1965) . IgD IgE 16 IgG, the main immunoglobulin found in the serum, makes up 75-85% of the total serum immunoglobulins. It is synthesized during a secondary immune re-sponse. IgG is a monomer in both the membrane bound and secreted forms. The x chain constant region has 3 domains compared to 4 for the u chain. In pigs two isotypes of IgG have been identified: IgGi and IgG2 (Pery 1973). IgGi is re-sistant to papain digestion and does not fix complement. IgG2 is cleaved by papain and is effective at fixing complement. A third isotype of porcine IgG has been tentatively been identified using monoclonal antibodies but has not been characterized (Pescovitz et a l . 1985.). IgA is the main immunoglobulin class found in secretions and on body surfaces. It occurs as a monomer in the membrane bound form and a dimer in the secreted form (see Figure 8). The two monomers are joined by the J chain as f o r IgM. In addition IgA synthesized in the mammary gland i s covalently linked to a glycoprotein called the secretory component. The secretory component increases the resistance of IgA to proteolytic enzymes in the small intestine (Tomasi and Calvanico 1968). In pigs IgA is in high concentration in sow's milk and i n the small intestine (Allen and Porter 1977) and is therefore of great importance to the survival of the neonatal piglet. Only these 3 classes of immunoglobulins have been characterized i n pigs. IgE and IgD also exist in other mammals including mice, rats and man,. IgE is responsible for immediate hypersensitivity reactions in allergy. IgE binds to Fc receptors on mast ce l l s . The binding of certain types of antigens causes the mast c e l l to release vasoactive molecules that produce the symptoms of allergy (Cooper et a l . 1984). IgD is found only in a membrane bound form. I t occurs on the membrane of unprimed B cells and i s a membrane receptor f o r antigen (Fu et a l . 1975). 17 The ontogeny of murine B c e l l s i s shown i n Figure 4. Stem c e l l s d i f f e r e n t i a t e i n t o pre-B c e l l s . I n i t i a l l y the pre-B c e l l s do not express mem-brane-bound immunoglobulins (Landreth et a l . 1981). The f i r s t event i n the production of immunoglobulins i s the rearrangement of the 4 genes that code f o r Figure 4. The ontogeny of B c e l l s (Based on Cooper et a l . 1984). 18 the u chain. They are the V, D, J and C genes (Ravetch et a l . 1 9 8 1 ) (see Figure 5). The V, D and J genes code for the variable domain of the heavy chain. There are many forms of each of these genes. During maturation of the lymphocyte the genes undergo translocation so that one V, D and J gene l i e together with a l l the intervening genes having been excised. The combining of the d i f f e r e n t V, D and J genes leads to a large number of possible amino acid sequences i n the*variable region of the immunoglobulin molecule. In addition, this trans-l o c a t i o n i s extremely prone to errors so that more v a r i a b i l i t y i s generated. Figure 5. The genetic basis of antibody production (Based on Tizard 1 9 8 6 b ) . Light Chain Heavy Chain Vi v2 V 3 J l J 2 C Variable Constant Germ l i n e DNA I Vi v2 V 3 J l J 2 Di D2 M G E A Lymphocyte DNA 1 V J D M G E A IgG Production Lymphocyte RNA 1 V J D G Variable Constant Protein Complete antibody 19 About 9,600 different gene arrangements of the heavy chain genes are possible in mice (Max 1984). Another 800 possible arrangements are possible for the light chains giving nearly 8,000,000 different spe c i f i c i t i e s of antibody i n mice. There are C genes for a l l of the different classes of immunoglobulins. I n i t i a l l y only the u gene is expressed and u chains appear in the cytosol of the pre-B c e l l . A similar process occurs for the rearrangement of the lambda and kappa genes. The kappa gene is rearranged f i r s t (Korsmeyer 1981). If the rearrangement is successful then the lambda gene is not rearranged. If i t i s unsuccessful the lambda gene is rearranged. Only one of the lambda or kappa chains is produced by any given B c e l l . Once the light chain has been rear-ranged, IgM is produced by the B c e l l and expressed on i t s membrane. IgD is also produced and occurs on the membrane of the mature, virgin B c e l l (Fu et a l . 1975). When mature, virgin B cells are activated by T helper cells, they begin to divide and differentiate into two types of cells (Cooper et a l . 1984). Plasma cells lose their surface immunoglobulin and begin to produce secretory type IgM. This primary response takes 10-14 days to develop. It is the only type of response the neonatal piglet is capable of mounting. A small number of the dividing cells differentiate into memory B c e l l s . These cells are long lived. The memory cells also undergo a switch in the type of immunoglobulin they produce. They w i l l have the same variable regions as they started with but now they may produce IgG or IgA depending on type of antigen, the location of the immune response and other poorly understood factors. The next time the antigen that triggered the response is met, the number of B c e l l s able to produce antibodies to i t wi l l be greatly expanded and the response w i l l be IgG or IgA. This is termed a secondary response. 20 B cells are the last link in the immune response. The macrophages ingest, process and present antigen bound to class II MHC molecules. An antigen specific T helper c e l l binds to the antigen-class II complex causing the release of interleukin-1. The T c e l l clone is then activated. It in turn activates antigen specific B cells by c e l l - c e l l interactions and the release of activating fac-tors. The B c e l l begins to divide and form plasma c e l l s . The plasma cells produce IgM for a primary response and IgA or IgG for a secondary response. B cells also differentiate into memory cells to enhance the immune response i f the same antigen is encountered again in the future. When the foreign antigen has been eliminated, suppressor T cells stop the immune response by acting upon T helper c e l l s . C o m p l e m e n t The complement cascade is a group of about 20 proteins found in the blood-stream of a l l vertebrates. When the precursors of the complement cascade are activated, other complement components are activated sequentially in a cascade. The blood-clotting system is another example of a biological cascade system. Complement is not part of the specific immune system, but interacts closely with many components of i t . The main functions of the complement cascade are: (1) production of inflammation; (2) opsonization of antigens for phagocytosis and (3) cytotoxicity towards bacteria (Brown et a l . 1984). Cytotoxicity against bacteria i s the most obvious function of complement and can be activated by two pathways. The classical pathway is initiated by the binding of antibody to antigen (see Figure 6). IgM is more effective at ini t i a t i n g the classical pathway than IgG (Borsos et a l . 1981). This is because two immunoglobulin molecules must be bound to antigen side by side. The pent-americ structure of IgM makes this far more lik e l y than for monomeric IgG. IgA 21 IgA w i l l not i n i t i a t e the complement cascade and may block i n i t i a t i o n by IgG (Grifiss 1975). The binding of IgM or IgG to antigen changes the 3 dimensional structure of the Fc portion of the molecule. This allows the CI Figure 6. The classical, alternate and terminal pathways of the complement cascade (Based on Tizard 1986c). Antigen component cf ccj.plarr.ent to bind to the immunoglobulin molecules. This a c t i v a t e s CI which cleaves two other complement p r o t e i n s . C4 i s converted to C4a and C2 i s converted to C2b. These p r o t e i n s then combine to form C4a2b which cleaves C3 to C3a and C3b (Tizard 1986c) . The te r m i n a l complement pathway i s i n i t i a t e d by C3b (see Figure 6) . C3b cleaves C5 i n t o C5a and C5b. C5b then combines with C6, C7 and C3 to form a complex. This complex serves as a seed to polymerize the l a s t component C9. Twelve to f i f t e e n C9 molecules j o i n together to form a tubular s t r u c t u r e . This s t r u c t u r e then i n s e r t s i t s e l f i n t o the membrane of the b a c t e r i a that the immunoglobulin molecules were bound to, lea d i n g to osmetic l y s i s of the bac-t e r i a l c e l l . Complement can al s o be a c t i v a t e d i n the absence of immunoglobulins by the a l t e r n a t e pathway (see Figure 6). The C3 molecule w i l l spontaneously breakdown and i n i t i a t e t h i s pathway. This could cause considerable damage to host t i s s u e s so i t i s t i g h t l y regulated by f a c t o r H which i n h i b i t s the alternate-pathway. Factor H i s present i n high concentrations on the surfaces of host c e l l s . I t i s bound by s i a l i c a c i d which a l s o occurs i n high concentrations or. the surfaces of host c e l l s (Katzatchkine et a l 1979). B a c t e r i a however have low concentrations of s i a l i c a c i d on t h e i r surfaces and therefore lew concen-t r a t i o n s of f a c t o r H. The a l t e r n a t e pathway can then proceed, l e a d i n g t c the l y s i s of the b a c t e r i a . Some components of complement have other f u n c t i o n s i n a d d i t i o n to t h e a c t i v a t i o n of the cascade. C3b binds to b a c t e r i a l antigens and functions as an opsonin f o r macrophages and n e u t r o p h i l s . I t also promotes the production of immunoglobulins by B c e l l s . C5a promotes the chemotaxis cf phagocytic c e l l s . 23 C3a and C5a provoke the inflammatory repsonse by causing mast c s l l degrar.ulatior. (Brown et a l . 1934) . D e f i c i e n c i e s i n complement lead to recurrent b a c t e r i a l i n f e c t i o n s (Tizard 1936c). Complement i s an important e f f e c t o r mechanism of the immune system. I t i s a l s o important i n i t s own r i g h t as a component of n o n - s p e c i f i c host defense. The Ontogeny of the P i g l e t ' s Immune System The immune system of the p i g l e t develops i n a s e r i e s of w e l l defined steps during the 115 day g e s t a t i o n p e r i o d (see Table 3). Stem c e l l s i n the yolk sac begin e r y t h r c p o i e s i s and lymphopoiesis on days 15-16. The thymus begins to develop around day 21 ( S t e r z l and S i l v e r s t e i n 1967) . Stem c e l l s from the yolk sac c o l o n i s e the thymus cn Day 23 and t h i s begins the process of T lymphocyte maturation. The percentage of f e t a l p i g lymphocytes with surface immunoglobulins i s lew p r i o r to 70 days of g e s t a t i o n . The number of I g + c e l l s increases r a p i d l y from 70 to 30 days. A s i m i l a r p a t t e r n i s seen i n serum immunoglobulins. On day 50 of g e s t a t i o n , the c i r c u l a t i n g l e v e l s of IgG are 2.5 ug mLr1 . By day SO t h i s increases to. 12.1 ug rnL-1 (Franz et a l . 1982). These values are extremely lew compared to adult serum IgG values of 10-15 mg rnL-1 (Jonnson 1973). IgG i s the predominant immunoglobulin i n the serum and amniotic f l u i d while IgM i s i n highest concentration i n the f e t a l l i v e r , i n t e s t i n e , spleen and thymus (Franz et a l . 1932). The period from 72 to 83 days seems to be the period during which the f e t a l p i g l e t develops immunocompetence (Solomon 1971) . Before t h i s period there are few p e r i p h e r a l T c e l l s and the i n j e c t i o n of a l l o g e n e i c lymphoid c e l l s does not provoke an immune response. From 72 to S3 days ges-24 t a t i o n , the number of p e r i p h e r a l T c e l l s increases massively. At 80 days of g e s t a t i o n , i n j e c t i o n of a l l o g e n e i c lymphoid c e l l s provokes a strong immune response. The f e t a l p i g l e t i s a l s o able to mount an immune response to sheep erythrocytes (Solomon 1971) on day 80 of g e s t a t i o n . Once the p i g l e t can respond to a given antigen, i t s response appears to be a d u l t - l i k e with respect to both humoral and c e l l u l a r immune components (Solomon 1971) . T a b l e 3. Ontogeny of the immune system of the f e t a l p i g l e t . G e station Dav Event Reference 21 E p i t h e l i a l thymus rudiment Mendel et a l . 1977 22 Spleen rudiment appears Mendel et a l . 1977 28 Lymphocytes i n thymus Mendel et a l . 1977 38 Lymphoblasts i n blood Solomon 1971 48 Lymphocytopoiesis i n Spleen Mendel et a l . 1977 50 Peyer's patches rudiment Chapman et a l . 1974 50 IgM, IgG and IgA found i n serum Franz et a l . 1982 52 Lymphocytopoiesis i n lymph nodes Mendel et a l . 1977 74 Onset of t r a n s p l a n t a t i o n immunity Solomon 1971 77 Lymphopoiesis i n Peyer's patches Solomon 1971 80 S e n s i t i z a t i o n to a l l o g e n e i c lymphoid c e l l s Solomon 1971 80 Antibody response to Sheep erythrocytes Solomon 1971 25 Host Defenses in the Neonatal Piglet The immune system of the neonatal p i g l e t i s capable of responding to antigens but i s unprimed. The primary immune response occurs too slo w l y to provide much p r o t e c t i o n against pathogens that the p i g l e t encounters on l e a v i n g the uterus. The p i g l e t must therefore r e l y on phagocytic c e l l s such as n e u t r o p h i l s and macrophages and on a n t i b a c t e r i a l f a c t o r s such as complement, l a c t o f e r r i n and lysozyme. Neu t r o p h i l s occur i n greater numbers i n the neonatal p i g l e t than i n adults (Sellwood et a l . 1986). The n e u t r o p h i l s of the neonate are as e f f e c t i v e at k i l l i n g E. c o l i as those from adults but because the neonate lac k s immunoglob-u l i n s to act as opsonins, n e u t r o p h i l b i n d i n g to b a c t e r i a i s impaired. C o l o s t r a l and milk immunoglobulins provide a source of opsonins and enhance k i l l i n g by n e u t r o p h i l s i n calves (Renshaw et a l . 1976). The presence of s p e c i f i c immuno-g l o b u l i n s i n the i n t e s t i n e a l s o has an e f f e c t on the emigration of n e u t r o p h i l s i n t o the i n t e s t i n a l lumen (Sellwood et a l . 1986). For n e u t r o p h i l emigration to take place, there must be chemotactic s i g n a l s from damaged t i s s u e and s p e c i f i c immunoglobulin must be present i n the lumen of the gut. In colostrum deprived p i g l e t s , and p i g l e t s r e c e i v i n g non-immune colostrum there i s no emigration of n e u t r o p h i l s i n response to a challenge to E. c o l i . Complement a c t i v i t y i s approximately one h a l f of adult l e v e l s i n neonatal p i g l e t s (Rice and Ecuyer 1963). Complement components are absorbed from c o l o -strum during the f i r s t day of l i f e and increase the p i g l e t s complement a c t i v i t y (Day et a l . 1969) . Complement a c t i v i t y reaches adult l e v e l s by 2-4 weeks a f t e r b i r t h . B c e l l numbers are about one t h i r d of the adult l e v e l s i n the neonatal p i g l e t (Banks 1981). Serum immunoglobulin concentrations are more d i f f i c u l t to d i s c e r n 26 because of the presence of passively absorbed and actively synthesized immunoglobulins in naturally reared piglets. In piglets a r t i f i c i a l l y reared on bovine colostrum the amount of actively synthesized immunoglobulin can be determined. Serum IgM increases rapidly during the f i r s t week after birth and remains the predominant serum immunoglobulin until approximately 21 days of age (Klobassa et a l . 1981). Serum IgG increases slowly until day 21. It then increases rapidly and becomes the predominant immunoglobulin in the serum. The number of B cells in the intestinal mucosa of the neonatal piglet is extremely low. The colonization of the intestinal mucosa by B cells begins in the duodenum and the number of B cells is higher in the duodenum than in other portions of the gut throughout development (Allen and Porter 1977). IgM bearing cells are the f i r s t to appear. IgA bearing cells are in the minority until 4 weeks of age. By 12 weeks of age, 90% of the B cells of the intestinal mucosa bear surface IgA. The T c e l l function of neonatal piglets is different from adults in two ways. T c e l l numbers in the peripheral blood and in the small intestine of neonatal piglets are low (Cepica and Derbyshire 1984a; Chu et a l . 1979) and the relative numbers of the different T c e l l populations are different from adult pigs. Peripheral blood T cells of neonatal piglets have a suppressive effect on the differentiation of B cells into immunoglobulin secreting cells (Suganuma et a l . 1986). At 6 weeks of age the suppressive effect disappears and B c e l l levels increase to approximately one half of adult levels. No sig-nificant T helper c e l l function i s present at 6 weeks of age. There are also fewer cytotoxic T cells in the peripheral blood and small intestine of newborns (Cepica and Derbyshire 1984a; 1984b). T cells, from newborn piglets show no cytotoxic activity against transmissible gastroenteritis (TGE) virus infected 27 c e l l s . Low lymphocyte numbers, direct suppression and lack of effector and helper T cells a l l account for the susceptibility of neonatal piglets to TGE and other v i r a l infections. The age of weaning has a significant effect on both cell-mediated and antibody responses in the piglet. Weaning prior to 5 weeks of age resulted in impaired c e l l mediated immunity (Blecha et a l . 1983). Weaning at 4 days of age followed by a r t i f i c i a l rearing on immunoglobulin-free milk replacer resulted in decreased antibody responses to sheep red blood cells or Salmonella antigen (Haye and Kornegay 1979). The factors in sow's milk responsible for these effects have not been determined. In summary, the neonatal piglet is extremely immunodeficient. It has virtually no circulating immunoglobulins. It does have high numbers of neu-trophils and macrophages but the lack of immunoglobulins to act as opsonins decreases their a b i l i t y to phagocytize foreign particles. Low levels of serum complement increase the susceptibility to bacterial infections. Lymphocyte numbers are low. B c e l l and T effector c e l l populations are reduced while T suppressor c e l l activity is increased. For survival, the piglet is almost entirely dependent on sow's milk and colostrum to augment and regulate i t s immune system. 28 PROTECTIVE FACTORS IN SOW'S COLOSTRUM AND MILK Sow's milk and colostrum provide nutrition and immune protection to the neonatal piglet. Piglets are born with very low energy reserves. Glycogen stores are about 65 Kcal while fat stores add another 100 Kcal (Aumaitre and Seve 1978). Without an exogenous source of energy soon after birth, piglets rapidly become hypoglycemic, hypothermic and die. Piglets also have limited stores of protein, vitamins A, C and B complex and iron. Since nutritional deficiencies increase host susceptibility to disease (Beisel 1984), i t is extremely important to provide a balanced diet to the piglet immediately following birth. Early work with the a r t i f i c i a l rearing of piglets concentrated on the composition of sow's milk and developing a substitute (Braude et a l . 1947) (see Table 4). Table 4 . Approximate chemical analysis of sow's colostrum and milk (Aumaitre and Seve 1978). component g/100 mL of: Colostrum Milk  Dry Matter 23.0 18.3 Fat 4.7 6.6 Protein (N x 6.38) 13.8 5.3 Lactose 3.6 5.5 Ash 0.7 0.8 Energy content (Kcal 100 mL-1) 140 107 29 The f i r s t attempts to rear colostrum deprived piglets in non-sterile environments using these sow milk replacers resulted in 100% mortality (Bustad et a l . 1947). Experiments using hysterectomy derived piglets reared in com-pletely germ free environments were more successful and survival rates of from 70 to 95% were obtained (Young and Underdahl 1953; Young et a l . 1955; Betts et al I960;). The expense and complexity of rearing germ free piglets made i t useless as a management tool to be used by producers. Obviously more than the nutritive properties of sow's milk and colostrum are important in ensuring piglet survival. The main problem is in looking at sow's milk and colostrum in the same manner as other feeds. Milk is not an amorphous liquid. It is a highly structured and complex system of selected proteins, l i p i d s , carbohydrates and other nutrients. Sow milk replacers must replace the functional properties of these components as well as their nutritional properties. As in the case of host defense mechanisms, the protective components of milk and colostrum can be divided into non-specific and specific elements. The non-specific protective components of milk and colostrum can be further subdivided into different functional groups. Nutrient binders such as the iron-binding protein lactoferrin and vitamin B12 binding proteins sequester nutrients in forms that make them unavailable for the growth of microorganisms in the intestinal tract. Complement components can destroy bacteria non-specifically or in concert with specific antibody. Milk enzymes either attack bacteria directly (lysozyme) or produce antibacterial products (lactoperoxidase). A whole range of non-immunoglobulin proteins that mimic enterocyte receptors for bacteria and viruses are present in milk. Bacteria and viruses that bind to these soluble receptor analogues are prevented from binding to the entero-cytes. 30 In addition to these non-specific protective components, sow's milk and colostrum also contain viable lymphocytes, macrophages, neutrophils and im-munoglobulins. These provide specific immune protection systemically and locally in the intestinal tract. Viable Cells Lymphocytes, macrophages and neutrophils are normal constituents of sow's milk and colostrum. These cells are not random collections of cells from the sow's own immune system. They are selected populations of cells derived from a compartment of the immune system known as the secretory immune system. The cells are specific for enteropathogenic organisms and food antigens that the sow has encountered previously. Since the piglets are being reared in the same environment as the sow, i t is li k e l y that the piglet w i l l encounter the same bacteria, viruses and food antigens. The secretory immune system is the portion of the immune system that is concerned with immunity on the mucosal surfaces of the body. This includes the mucosal epithelia of the gastrointestinal tract, the mammary gland, the lungs, salivary glands, uterus and skin. Two things are typical of secretory immune responses: (1) cells involved in an immune response at one mucosal site migrate to other mucosal sites and provide immune protection there and (2) secretory IgA is synthesized during secretory immune responses. The portion of the secretory immune system that directly affects the survival of piglets is the gut-mammary axis of immunity. Gut-associated lyphoid tissue (GALT) consists of a covering of specialized epithelial cells called M cells overlying lymphoid tissue. The M cells take up intact macromolecules from the lumen of the gut via pinocytosis and transport them to the lyphocytes and macrophages in the underlying lymphoid tissue (Bockman and Cooper 1973). In human studies i t has been found that small amounts of intact food antigens are transported from GALT to the mammary gland and into the milk (Hemmings 1980). This exposes even exclusively breastfed infants to food antigens. The B c e l l population of GALT consists of IgA and IgM bearing c e l l s . Antigen stimulation of GALT leads to the activation of only IgA secreting plasma c e l l s . No IgG secreting B cells are produced. This effect i s caused by a population of "switch" T cells in GALT (Kawanishi et a l . 1983). These T cells cause undif-ferentiated IgM bearing B cells to differentiate into IgA secreting cells only. T cells that regulate the synthesis of IgA by committed B cells also exist. These T cells have Fc receptors for the a chain of IgA. They produce soluble help and suppression factors that bind to the surface IgA on B cells (Kiyono et a l . 1985; Hoover and Lynch 1983). These T cells regulate the synthesis of immunoglobulin by IgA secreting B cells but have no effect upon IgG synthesis. Upon stimulation by T cells, B cells present in GALT migrate to the mesen-teric lymph nodes, the blood, liver and spleen (McWilliams et a l . 1975). They then home back to the gut or other mucosal sites including the mammary gland (Phillips-Quagliata et a l . 1983). In a study with pregnant women, oral immunization with nonpathogenic E. c o l i resulted in the appearance of specific B cells in the colostrum of the volunteers (Goldblum et a l . 1975). In another human study, the transfer of cell-mediated immunity to colostrum after oral immunization was demonstrated (Parmely et a l . 1976). Bohl et a l . (1972), de-monstrated that production of IgA specific for TGE virus in the colostrum of sows depended on prior enteric infection with the virus. Presumably, migration of specific lymphocytes from the GALT to the mammary lymphoid tissue was re-sponsible for this mechanism. The population of IgA secreting B cells in the sow's mammary gland increases markedly 1 to 2 weeks prior to farrowing (Brown et a l . 1975). One study found that in mice this increase in cells around parturition was under hormonal control (Lamm et a l . 1977). Virgin female mice given injections of estrogen, progesterone and prolactin had greatly increased numbers of IgA secreting B cells in their mammary tissues. Treatment with testosterone decreased the numbers of these cells in the mammary gland. Total c e l l counts in sow's colostrum averaged 1 x 107 on the day of farrowing and decreased to 1 x 106 by 10 days after farrowing (Evans et a l . 1982). The differential c e l l counts from that study are shown in Table 5. Sow's milk and colostrum differs from human and bovine milks in that neutrophils are more numerous than macrophages (Evans et a l . 1982). Table 5. : Differential c e l l counts for the mammary secretions of sows in which bacterial infection was absent (Evans et a l . 1982) • Days post % of total c e l l vield partum neutrophils macrophages lvmphocvtes epithelials eosinophils <1 71.7 1.3 26.4 0.4 0.2 3 55.4 15.0 22.8 6.1 0.7 10 39.2 14.6 13.7 31.4 1.1 15 51.3 5.5 11.2 31.3 0.7 20 32.1 0.0 0.0 67.9 0.0 33 The lymphocytes that appear in milk and colostrum are derived from the secretory immune system and are different from lymphocytes found in serum. Most of the B cells are IgA secreting cells (Parmely and Beer 1977) . A high percentage of these B cells are specific for enteric microorganisms. The T c e l l populations of milk and colostrum also show different reactivities to antigens than peripheral blood T c e l l s . The role that these cells play in protection of the piglets against enteric infection has not been well explored. Colostral macrophages protect suckling rat pups against Klebsiella-induced necrotizing enterocolitis (Pitt et a l . 1974). In vitro studies have shown that neutrophils from human milk can phag-ocytize and k i l l E. c o l i (Robinson et a l . 1978). Various studies have also reported the transfer of c e l l mediated immunity via lymphocytes present in milk and colostrum. Head and Beer (1979) found that colostral lymphocytes in rat's milk could be absorbed directly into the bloodstream of suckling rats. Rat pups that were cross fostered to MHC incompatible mothers developed graft-versus host disease. Resistance to tumors can be transferred via milk in mice (Head and Beer 1979). When mice susceptible to a Leydig c e l l tumor were suckled by their natural mothers they had no resistance to challenge with tumor c e l l s . When the susceptible mice were suckled by resistant mothers they survived a challenge with tumor c e l l s . Since these types of experiments have not been performed with pigs, there is no information on whether the piglet benefits from ingesting the cells present in sow's milk and colostrum. Immunoglobulins Most investigations of the protective effect of sow's colostrum and milk have concentrated on their immunoglobulin content. There are several reasons 34 for this. Immunoglobulins make up a larger percentage of the total solids in colostrum and milk than other protective factors so they are the most obvious factor to start with. They are simple to isolate and purify so that invest-igating their properties is relatively easy. They are available in large quant-it i e s from sources such as abattoir blood, bovine milk and whey. This makes i t feasible to perform feeding experiments using a r t i f i c i a l diets f o r t i f i e d with purified immunoglobulins. The immunoglobulin content of sow's colostrum during the f i r s t day after parturition i s different from that found subsequently in the milk (Table 6). Table 6. Ig levels in serum, colostrum, milk and intestinal juice of pigs (Bourne 1973). Immunoglobulin concentration mg mL=-L IgG IcrA IcM IgArlgG Sow Serum 24.3 2.4 2.9 0.10 Colostrum (0 hr) 61.8 9.7 3.2 0.16 Milk (24 hr) 11.8 3.8 1.3 0.32 Milk (2 d) 8.2 2.7 1.8 0.33 Milk (3-7 d) 1.9 3.4 1.2 1.79 Milk (8-35 d) 1.4 3.0 0.9 2.14 Intestinal Juice 0.2 2.6 Trace 13.0 In pigs, IgG is present in high concentrations in colostrum while IgA pre-dominates in milk a few days after parturition. This pattern is not followed 35 by a l l species however. In primates and rodents, IgA is the principal immuno-globulin in both milk and colostrum. In ruminants, IgG predominates in both milk and colostrum. A l l the IgG and 40% of the IgA in the colostrum during the f i r s t few days after parturition are transferred from the blood stream of the sow and transferred directly into the colostrum (Bourne and Curtis 1973). In cows, specific receptors on the epithelial cells of the mammary gland transport IgG from the bloodstream to the milk (Brandon et a l . 1971). Selective receptors for immunoglobulin in the mammary glands of sows may also exist (Franek et a l . 1975). After parturition, the transfer of immunoglobulins from the blood to the milk is dramatically reduced. Three days after parturition, 90% of the IgA and 70% of the IgG in sow's milk comes from local production by B cells in the mammary gland (Bourne and Curtis 1973). This pattern of IgG in the colostrum and IgA in the milk appears to be adapted to providing optimal immune protection to the piglet. The immunoglob-ulins found in the colostrum are ingested by neonatal piglets and absorbed directly into the blood stream via pinocytosis. These immunoglobulins are therefore responsible for providing systemic immunity in the piglet. Since the majority of the immunoglobulins present in the colostrum are derived from the serum of the sow, they are ideally selected for this purpose. In addition IgG has a longer biological half l i f e (14 days) than either IgA (2.5 days) or IgM (5 days) (Bourne and Curtis 1973). IgG therefore provides much longer systemic protection than IgA or IgM. Since immunoglobulins are only absorbed into the blood stream during a short period after birth, i t is important that they should last as long as possible. Secretory IgA that is absorbed from colostrum has a 36 half l i f e of less than 24 hours (Bourne 1973). This i s not due to degradation of the slgA but rather i t s transport to mucosal surfaces where i t provides local immune protection (Bradley et a l . 1976). Serum immunoglobulin concentrations of conventionally reared piglets are shown in Figure 7. Figure 7 . IgG, IgA and IgM concentrations in sera of naturally reared piglets (n=24) (based on data from Klobassa et a l . 1981). Days 37 Absorption of macromolecules from the intestinal tract by piglets takes place in two distinct steps (Lecce 1973). The f i r s t i s internalization of macromolecules by pinocytosis. The second is transport of the internalized macromolecules across the enterocyte and into the blood stream. In rodents, a wide variety of macromolecules are internalized via pinocytosis but only IgG is transported to the blood stream. This selectivity does not exist in piglets. They w i l l internalize a wide range of macromolecules including dextrans and polyvinylpyrrolidone (Lecce et a l . 1961) but IgG is preferentially transported into the blood stream in the presence of other macromolecules (Leary and Lecce 1979). When porcine IgG or bovine albumin alone was fed to piglets, they were both absorbed at the same low level. When fed together, however, the absorption of IgG increased significantly while the bovine albumin was absorbed at the original low level. Leary and Lecce (1979) proposed that specific receptors on enterocytes bind IgG. Non-immunoglobulin macromolecules increase absorption of IgG by non-specifically stimulating pinocytosis. The non-immunoglobulin molecules are internalized non-specifically and are therefore absorbed less eff i c i e n t l y . This model has not been proven. Cessation of absorption of macromolecules directly into the blood stream (closure) usually occurs between 24 and 48 hours after birth. The ingestion of nutrients by the piglet seems to be the inducer of closure (Patt 1977). Starvation of piglets prolongs the period before closure (Lecce 1973). After closure occurs, milk provides protection against infections in the intestinal tract of the piglet. Sow's milk is rich in secretory IgA. The properties of this immunoglobulin are ideally suited to providing local immune protection. The structure of human IgA is shown in Figure 8. The monomeric subunits 38 Figure 8. The covalent structure of human monomeric, dimeric and secretory IgA (Based on Underdown and Schiff 1986). IgA l o n o i e r Secretory coiponent Kacosal t r a n s p o r t receptor p r o t e o l y t i c a l l y l a b i l e segient l l t t t U t U l i O t t <<HMIimtt<UtMHH<| MIM|ltl lMM>M«M«Wltll tMUMUM«r 39 of slgA have the a b i l i t y to polymerize. The heavy chain of the IgA monomer has a C terminal extension containing an extra cysteine (Koshland 1985). These cysteine residues allow the linking of 2 or more IgA monomers via the J chain. Dimeric IgA i s the most common form but trimers, tetramers and pentamers are also formed (Radl et a l . 1973). This polymerization step i s essential for binding of the IgA to the mucosal transport receptor. IgA is secreted into the lymphoid tissue where i t is synthesized. To provide protection on mucosal surfaces i t must then be actively transported through the mucosal epithelia and released. The transport of IgA from blood to bile in perfused livers has been used to study the mechanism of this transport system (Fisher et a l . 1979). Secretory component is synthesized in the interior of the c e l l . It consists of the polypeptide found associated with free slgA plus an extra segment (Mostov et a l . 1979). This extra segment inserts in the c e l l membrane and anchors the secretory component on the surface of the c e l l . IgA then binds to the membrane bound secretory component (Weicker and Underdown 1975). The entire complex then migrates across the c e l l where the membrane anchoring segment is cleaved off and the slgA i s secreted (Solari and Kraehenbuhl 1984). This transport system is constituitive (Mullock et a l . 1980) and secretory component is synthesized and moves across the epithelial cells whether IgA is present for transport or not. The energy expenditure for this is enormous. Thirty micrograms of secr-etory component per gram of tissue must be synthesized every hour to maintain the transport system (Underdown and Schiff 1986). Such a high expenditure of energy indicates the importance of slgA in the protection of mucosal surfaces. Secretory IgA must survive degradation in the digestive tract i f i t is to perform i t s protective function. The hinge regions of IgG and IgM molecules in 40 particular are susceptible to degradation by digestive and bacterial pro-teinases. The hinge region of IgA is resistant to proteolysis and therefore better adapted to performing i t s protective functions in the intestinal tract of the piglet than IgG or IgM (Plaut 1983). The protective functions of slgA are a l l related to i t s a b i l i t y to spec-i f i c a l l y bind to antigen. Secretory IgA in ingested milk causes the aggregation of microorganisms in the digestive tract of the piglet. The dimeric structure of slgA increases i t s abi l i t y to agglutinate particles compared to IgG but i t is inferior to pentameric IgM in this regard (Ishizaka et a l . 1965). Underdown and Schiff (1986) suggested that the structure of slgA is a compromise between the agglutinating properties of IgM and the tissue permeability of IgG. In addition to agglutination, slgA binds with cysteine residues in the mucous coat on intestinal epithelium. This provides a long lasting coating of specific antibody on the mucosal surface of the small intestine. This coating of slgA inhibits adherence of microorganisms to enterocytes (Walker and Isselbacher 1974) . IgG does not form this protective coating. Microorganisms that do manage to adhere to the specific receptors on enterocytes are prevented from being internalized by slgA. IgG does not prevent internalization (Underdown and Schiff 1986). The protective functions of IgA do not extend to k i l l i n g microorganisms. Rather i t functions in immobilizing and preventing adherence and internaliza-tion. Non-specific protective factors such as iron-binding proteins, lysozyme, lactoperoxidase and complement contribute to the k i l l i n g of microorganisms. In addition to binding microorganisms, slgA can also bind and neutralize toxins produced by pathogenic bacteria (Dobrescu and Huygelen 1976; Pierce and Reynolds 1975) . 41 Immunoglobulin Fortified Milk Replacers The i n i t i a l failures in rearing colostrum deprived piglets led to attempts to simulate the immune properties of sow's colostrum and milk by fortifying milk replacers with immunoglobulins. Obviously the model for doing this is the natural immunoglobulin content of colostrum and milk. The problem is obtaining a rich source of porcine slgA. The only practical source of porcine immuno-globulins is pig abattoir blood which is rich in IgG and poor in slgA (see Table 6). Fortification of milk replacers with porcine blood immunoglobulins results in a milk replacer immunoglobulin profile similar to sow's colostrum but lacking the slgA found in sow's milk. In vivo experiments using porcine blood immuno-globulins show that this lack of slgA does not have a detrimental effect on •survival (Owen and Bell 1964; Scoot et a l . 1972; McCallum et a l . 1977). Ammonium sulphate fractionated porcine blood immunoglobulins fed at a level of 10 g Kg body weight - 1 on Day 1 and 2 g Kg body weight - 1 on Days 2-10 gave survival rates of 70% (McCallum 1977). When immunoglobulins were given orally only during the f i r s t 24 hours of l i f e , the survival rates were very low (Owen et a l . 1961) . Porcine blood IgG provides adequate local protection in the intestinal tract of the piglet. Large scale production of immunoglobulins from porcine blood is possible using a combination of different purification procedures. Ammonium sulphate fractionation is the most common method in use. E l l i o t et a l . (1987) described a continuous process for the large scale purification of immunoglobulin from blood. Continuous desludging centrifuges are used to remove the cellular and f i b r i n components of blood. Saturated ammonium sulphate is added to serum to give 45% saturation. The immunoglobulin containing precipitate is collected by 42 decanting centrifugation and ammonium sulphate is removed by electrodialysis or u l t r a f i l t r a t i o n . The remaining water is removed by spray drying. This frac-tionation method has several disadvantages. The immunoglobulin containing fraction changes phase making resolubilization necessary. Also, extensive processing using u l t r a f i l t r a t i o n or electrodialysis is required to remove the ammonium sulphate. A more recent method of fractionating serum immunoglobulins was described by Lee et a l . (1988). This method uses polyphosphates to fractionate serum immunoglobulins. Only 1.04% polyphosphate is required and the immunoglobulin containing fraction does not change phase. This method is more readily adapted to continuous production of immunoglobulins than ammonium sulphate fractiona-tion. In addition to porcine blood, bovine blood, milk, whey and colostrum are potential sources of immunoglobulins for the a r t i f i c i a l rearing of piglets. Senft and Klobassa (1971) obtained survival rates of 92% feeding bovine colo-strum to 232 piglets in a non-isolated environment. The feeding of ammonium sulphate fractionated bovine blood immunoglobulins to a r t i f i c i a l l y reared piglets resulted in 100% survival for a group of 8 piglets (McCallum 1977). A direct comparison of bovine and porcine serum immunoglobulins on piglet per-formance has not been made however. One thing that a l l in vivo experiments have in common is that they use immunoglobulins as the sole protective component in the milk replacer. As mentioned previously immunoglobulins do not directly k i l l bacteria. Other non-specific factors in milk which act in concert with immunoglobulins, are responsible for bacteriostasis and the direct k i l l i n g of bacteria. 43 Lactoferrin Lactoferrin is an iron-binding single chain glycoprotein unique to mammals. It shares close homology with transferrin and ovotransferrin but differs in distribution in the body and in function. Transferrin occurs mainly in serum as a transport protein for ionic iron. Lactoferrin occurs in milk, seminal fl u i d , tears and on body surfaces in general. Its distribution i s similar to IgA and this reflects i t s functions as an important part of the immune response. Lactoferrin binds ionic iron, making i t unavailable to bacteria and thus pre-venting their growth. Lactoferrin is the predominant iron-binding protein in the colostrum and milk of commercially important agricultural species. For this reason there is interest in i t s use as a feed additive for milk replacers and starter diets. Lactoferrin was f i r s t isolated by Sorensen and Sorensen in 1939 (Groves, 1960) as a red protein from milk. Polis and Shmukler (1953) obtained this red protein in an impure form and were able to characterize i t as containing iron. The most common method of purification presently in use was developed by Jo-hanssen (1969) (Figure 9) . Lactoferrin tends to form complexes with other molecules such as spermatozoal surface components, DNA, serum albumin and fi-lactoglobulin. It also forms complexes with other lactoferrin molecules in the presence of Ca + + (Bez-korovainy, 1980). This makes i t d i f f i c u l t to prepare in a pure monomeric form. For this reason, estimates of the molecular weight of lactoferrin vary from 77-93 Kilodaltons (Groves, 1960; Castellino et a l . 1970; Weiner and Szuchet, 1975). Weiner and Szuchet (1975) reported the molecular weight of bovine lactoferrin as 93 ± 2 Kilodaltons. This latter value is probably the most reliable because of the conditions under which i t was obtained. The protein was shown 44 Figure 9. The preparation or bovine lactoferrin from milk (Based on Johanssen 1969). Fat discarded Supernatant discarded Lactoferrin-free — fractions discarded Lactoferrin-free — fractions discarded Milk Centrifuge 8000 x g for 40 min Skim milk pH 7.0 in 0.05 M K H 2 P O 4 / K 2 H P O 4 buffer Add 6 g CM-Sephadex C50 per L of milk Stir 2 hours Wash CM-Sephadex C50 with deionized water Apply to column Elute with linear NaCl gradient 0.2-0.5 M in pH 7.0 in 0.05 M K H 2 P O 4 / K 2 H P O 4 buffer Lactoferrin enriched fraction Gel f i l t r a t i o n with Sephadex G-200 Lactoferrin to be homogeneous and the same molecular weight was obtained by both Svedberg and high speed sedimentation techniques. The pi of lactoferrin was measured in the same study as 8.0 ± 0.2. Lactoferrin exists in two forms, iron free and iron saturated. The forms have quite different properties. Iron-saturated lactoferrin is more resistant to denaturation than iron-free lactoferrin ( Baer et a l . 1979). Purified iron-free lactoferrin was denatured at temperatures as low as 45 °C while iron-saturated lactoferrin remained stable up to 60 °C. They also found that the 45 a b i l i t y of iron-saturated lactoferrin to bind iron was partially independent of the degree of denaturation. That i s , the iron binding site remained able to bind iron after other portions of the protein had been denatured. The effect of low temperature storage on lactoferrin was studied by Goldsmith et a l . (1983). Storage of human milk samples at -20 °C for 4 weeks decreased the iron binding capacity of the milk lactoferrin by 30%. The primary structure of bovine lactoferrin has been only partially elu-cidated. More work has been done on human lactoferrin and w i l l be discussed here. Human lactoferrin shares homology with transferrin and hen ovotransferrin. In addition, lactoferrin shows internal homology. The amino acid residues of the N-terminal half of the lactoferrin molecule (residues 1-336) show marked homology with those of the C-terminal half (residues 337-679) of lactoferrin (Brock, 1985). This homology probably arose as a result of gene duplication during evolution. Each homology region has one iron binding site which are designated the N-terminal site and the C-terminal site. The two carbohydrate chains on the lactoferrin molecule are located on the C-terminal region and are not involved in iron binding (MacGillvray et a l . 1977). Lactoferrin w i l l bind iron at a pH as low as 2.0 compared to 4.6 for trans-ferrin (Van Snick et a l . 1973). This may reflect the fact that lactoferrin must bind iron in the acid environment of the stomach. The binding of two atoms of iron to lactoferrin may be represented as: HeLf + 2Fe 3 + + 2 HC03- <-> Fe 2Lf(HC0 3 )2 +6H+ The binding of a metal ion requires the simultaneous binding of an anion usually carbonate (Bates and Schlabach, 1973). Other anions such as EDTA, oxalate and nitrilotriacetate can replace bicarbonate (Schlabach and Bates, 1975). Arginine is the amino acid residue that binds the anion. 46 The a b i l i t y of lactoferrin to bind iron is reduced by the presence of citrate (Reiter et a l . 1975). Citrate w i l l remove iron from lactoferrin and the iron citrate is actively taken up by bacteria (Rosenberg and Young, 1974). Iron has an octahedral ligand f i e l d and so requires 6 ligands to bind i t effectively. The carbonate anion is the f i r s t of these ligands. The other amino acid residues involved in the iron-binding site have been studied in several ways. Teuwisson et a l . (1972) used hydrogen ion titration of lactoferrin in i t s iron-free and iron-saturated forms to study the problem. Between pH 7 and 10, the binding of one Fe 3 + ion by lactoferrin liberates 3 protons. The possible ligands are thiol , phenol guanidinium, ammonium and tryptophan. The optical spectrum or Fe2Lf resembles iron complexes with phenol so i t is probable that tyrosine is a ligand. The presence of tyrosine as a ligand in the. iron binding site was confirmed by circular dichroism (CD) (Mazurier et a l . 1976), proton magnetic resonance (Woodworth et a l . 1970) and ultraviolet difference spectroscopy (Krysteva et a l . 1976) . Electron paramagnetic resonance spectra of lactoferrin-copper chelates are similar to those of copper-nitrogen compounds (Aasa et a l . 1963) at pH 7.5. At this pH, imadazole nitrogen is the only group that would remain unprotonated. This indicates that histidine is present at the iron binding site. Chemical modification of histidine with diethylpyrocarbonate showed there is only 1 histidine residue for the 2 iron atoms bound in bovine lactoferrin (Krysteva et a l . 1975). This suggest the iron binding sites are different from each other. Tryptophan may also be a ligand since CD spectra show bands due to restricted rotation of tryptophan change when iron is bound. There is evidence that suggests that although the conformation of tryptophan residues do change during iron binding, they play no role as a ligand. Sulphenylation of tryptophan 47 residues of lactoferrin did not change i t s a b i l i t y to bind iron (Ford-Hutchinson and Perkins 1972). Precise information on the nature of the residues involved in binding iron in lactoferrin w i l l have to wait until the complete primary structure of lacto-ferrin has been determined and the iron-binding residues identified within the structure. There is some controversy over whether the two iron-binding sites are equiv-alent. Warner and Weber (1953) found strong cooperativity between the sites in conalbumin. That is the binding of iron at one site increased the a f f i n i t y for iron at the remaining site. Aasa et a l . (1963), found no difference in the iron-binding sites in transferrin using electron paramagnetic resonance spectra. Mazurier et a l . (1976), using the same technique, found that the sites were significantly different. Luk (1971) showed that transferrin w i l l bind only one ion of the large lanthanide ions Nd 3 + or P r 3 + but 2 of the smaller lanthanides Tb 3 + or Ho 3 +. More recently the iron-binding constants for each site was measured by equilibrium dialysis. The results showed that the C terminal site has a higher a f f i n i t y for iron than the N terminal site (Aisen et a l . 1978). This conclusively demonstrated the non-equivalence of the sites. Iron plays a central role in the battle between host and bacterial pathogens. That iron is the nutrient that occupies this role is not surprising when i t s properties are examined. The chemistry of iron in aqueous systems is dominated by i t s low solubility. Under physiological conditions iron occurs in the ferric (Fe 3 +) form which forms insoluble polymeric hydroxides. These compounds have a solubility product of l C r 3 7 (Weinberg, 1974). In addition to i t s low solubility free iron causes the formation of peroxides and hydroxyl radicals. For these reasons free iron is rare in living systems and is inevitably bound 48 by proteins or heme. This makes i t relatively easy for the mammalian hosts to limit iron availability to bacteria. Transferrin is the primary iron transport protein in the plasma. The binding constant of transferrin for iron is about 10 3 6 (Aasa et a l . 1963). This means that the concentration of free iron in the plasma is about 10- 1 8 M. Since most bacteria require iron at a concentration of about 10- 1 0 M, (Kochan, 1977) this concentration of iron is about 108 times too low to support bacterial growth. On body surfaces, in the mammary gland and in milk, lactoferrin has the same effect as transferrin. It has a binding constant of about 10 3 6 for iron. Many bacteria are unable to grow in the presence of iron-free, lactoferrin. In vitro studies of the antibacterial effects of lactoferrin show that i t inhibits the growth of a wide variety of bacteria. Rainard (1986a), compared the susceptibility of various species of mastitis causing bacteria to lacto-ferrin. A l l 35 strains of E. c o l i were susceptible to inhibition by lactoferrin. Four of 10 strains of Staphylococcus aureus showed resistance to lactoferrin induced bacteriostasis. A l l strains of Streptococcus aqalactiae and Str. uberis were unaffected by the presence of lactoferrin. Rainard (1986a) observed that the iron requirements of the latter two organisms are low. This gives them re-sistance to inhibition by lactoferrin. Many other in vitro studies showed that lactoferrin inhibits the growth of E. c o l i (Rainard, 1986b; Reiter et a l . 1975; Samson et a l . 1979; Bullen et a l . 1972). For lactoferrin to prevent the growth of E. c o l i in the small intestine of the suckling animal i t must reach the small intestine intact. Lactoferrin is resistant to digestion by trypsin and chymotrypsin but not by pepsin (Brock et a l . 1976). Lactoferrin isolated from the feces of breast fed infants was s t i l l able to bind iron (Spik et a l . 1978) . Lactoferrin's a b i l i t y to bind iron at pH's 49 as low as 2.0 (Van Snick et a l . 1973) means that i t can bind iron in the acid environment of the stomach. Bovine milk contains 4-8 mM of citrate. Experiments using bovine milk showed no bacteriostatic effect due to lactoferrin unless citrate was removed by dialysis (Reiter et a l . 1975). Citrate is absorbed in the upper duodenum in calves (Reiter, 1978). Bicarbonate is also released in the duodenum. This gives a low citrate, high bicarbonate environment, ideal conditions for iron-binding by lactoferrin. There are few in vivo studies on the anti-bacterial properties of lacto-ferrin. Two in vivo models have been used, the bovine mammary gland and the suckling guinea pig. Lactoferrin concentrations in bovine mammary secretions are shown in Table 7. Seventeen lactating cows had one quarter infused with a pathogenic strain of E. c o l i . Thirteen out of 17 quarters became infected (Reiter, 1985a). During the dry period however, none of 14 quarters became infected. Two dry quarters infused with both iron and bacteria became infected. Clearly the high lactofe-r r i n levels in the dry secretion of the mammary gland provides protection from bacterial infections. The lower levels found during lactation along with the high citrate levels mean a greater susceptibility to infection. Bullen et a l . (1972) studied the effect of lactoferrin on the intestinal flora of suckling guinea pigs. Guinea pigs fed hematin as a source of iron had 10,000 times the numbers of E. c o l i than guinea pigs receiving no hematin had. For most virulent strains of E. c o l i this inhibition is temporary. This is because many these strains are capable of obtaining iron via another 50 Table 7. Concentration of lactoferrin in various secretions of the bovine mammary gland (Smith and Schanbacher, 1977). Type of mammary Lactoferrin secretion (mg mhzA.) Colostrum 1-5 Normal Milk 0.1-0.35 Early Involution 1-8 30 days of Involution 20-30 Cli n i c a l Mastitis 1-8 mechanism. Bacteria can obtain iron from the environment in two ways (Bullen et a l . 1978). The f i r s t i s by simple diffusion of free iron or iron compounds through the c e l l membrane. This is prevented in the presence of iron-free transferrin or lactoferrin. The second mechanism involves the synthesis by bacteria of iron chelators known as siderophores (Weinberg, 1984). In low iron environments many bacteria produce iron-chelators known as siderophores and membrane bound siderophore receptors. The ab i l i t y of bacteria to synthesize siderophores and their receptors i s a recognized virulence factor (Rogers, 1973). Siderophores are synthesized by many bacteria and secreted into the tissues of the host. They have iron binding constants of up to 10'2 and are able to remove iron from transferrin or lactoferrin. The siderophore receptors transport the iron-siderophore complex into the bacterial c e l l where enzymes remove the iron. To combat this, mammalian hosts further limit the availability of iron in 51 the body. Leukocytic endogenous mediator-endogenous pyrogen (LEM-EP) is secreted by leukocytes at the onset of a bacterial infection (Powanda and Beisel, 1982; Merriman et a l . 1977). This compound has three main effects on iron metabolism. The f i r s t is a shift of iron from the plasma to storage forms in the liver and spleen. Neutrophils secrete iron-free lactoferrin at the site of the infection. The lactoferrin removes iron from transferrin. Macrophages then remove the iron-saturated lactoferrin from the circulation. (Van Snick et a l . 1974). The macrophages transport the iron to the liver where i t is sequestered in storage forms. LEM-EP also causes a rise in body temperature. The synthesis of siderophores is inhibited by temperatures greater than 37 °C. The rise in body temperature that occurs during infections has the effect of suppressing synthesis of mi-crobial siderophores (Kluger and Rothenberg, 1979). Intestinal absorption of iron decreases during infection (Beresford et a l . 1971). LEM-EP may be responsible for this effect by increasing the production of f e r r i t i n in mucosal ce l l s . When the mucosal cells are shed this iron i s lost. (Sherman, 1984). In summary lactoferrin has two physiological roles. The f i r s t is to deprive bacteria of iron on body surfaces, the mammary gland and in milk and colostrum. The second is to scavenge iron from the body and return i t to the storage forms during bacterial infections. The lactoferrin content of sow's milk was measured in two studies. Masson and Heremans (1971) found that sow's milk contained 20-200 jig rnL-1 of lacto-ferrin and 20-200 ug rnL-1 of transferrin. E l l i o t et a l . (1984) measured the level of lactoferrin in sow's milk throughout a 21 day lactation. The levels ranged from 1100-1300 pg mL-1 for the f i r s t three days after farrowing. This dropped to 300 pg mL-1 on day 7 and 100 ug mL-1 on day 21. They noted that the 52 drop in lactoferrin matched the rise in immunocompetence of the piglet. Lactoferrin iron is well absorbed by piglets (Fransson et a l . 1983). 3 9 F e labelled lactoferrin was compared to 3 9FeSC-4 as a source of iron to piglets. The lactoferrin iron was absorbed at least as well as the FeSC-4. However, lactoferrin iron has the advantage of not being available to many species of bacterial pathogens. Synthetic Iron Chelators In spite of lactoferrin's desirable antibacterial properties, i t is far too expensive to use as an additive to sow milk replacers at the present time. The same is true of other iron-binding proteins such as transferrin and conalbumin. A l l of these proteins have relatively high molecular weights. Approximately 830 g of lactoferrin are required to bind 1 g of iron. Normally, lactoferrin is only about one third saturated with iron (Bezkorovainy 1980). To attain this level of saturation, 2,490 g of lactoferrin per g of iron are required. In addition to this, commercial milk replacers contain higher concentrations of iron than sow's milk even when no supplemental iron is added. This is due to contamination during processing and the iron content of the water used to reconstitute the milk replacer. The iron content of sow's milk i s 1 ng rnL-1 (N.R.C. 1978). The iron content of the reconstituted milk replacer used in the experiments described herein was 1.6 ug mL_1. The amount of lactoferrin required to bind this amount of iron at 33% saturation is approximately 4.0 mg mL_1. This is about 13 times the concentration of lactoferrin in sow's milk on Day 7 of lactation (Elliot et a l . 1984). Lactoferrin has no enzymic properties and i t s only biological role is as an iron chelator (Brock 1985). It may therefore be possible to replace lactoferrin with synthetic iron chelators. A number of synthetic iron chelators have been investigated for use in the 53 treatment of iron overload in humans (Grady and Jacobs 1981). Two compounds in particular appear promising. N,N'-ethylenebis-[2-(o-hydroxyphenyl)]-glycine (EHPG) was f i r s t synthesized by Frost et a l . (1958). This compound is also referred to as ethylene diamine-di-orthohydroxyphenyl acetic acid (EDDA) and this name w i l l be used here. Another iron chelator i s N,N*-bis(o-hydroxyben-zyl)-ethylenediamine diacetic acid (HBED). It was f i r s t synthesized by L'Eplattenier et a l . (1967). The structures of EDDA and HBED are shown in Figure 10. Compared to lactoferrin they have low molecular weights. Figure 10. The molecular structure of EDDA and HBED (L'Eplattenier et a l . 1967; Frost et a l . 1958). H B E D -o—CO H H OC—O-I I / C H x C H * . | I "oocc; CH,COO_ H H F e r r i c E D D A F e r r i c H B E D 54 EDDA has a molecular weight of 360.2 D and HBED has a molecular weight of 388.5 D. Both compounds are sexedentate ligands (L'Epplatenier et a l . 1967). This means that one molecule of either EDDA or HBED is required to bind one atom of iron. Compared to the 830 g of lactoferrin required to bind 1 g of iron, only 6.44 g of EDDA or 6.96 g of HBED are needed. EDDA and HBED are extremely specific for Fe 3 + (see Table 8). The af f i n i t y of HBED for Fe 3 + is 10 1 8- 3 times stronger than for the next most strongly bound metal ion Cu 2 +. In comparison the st a b i l i t y constant for lactoferrin bound iron is approximately 10 3 6 (Aasa et al 1963) . HBED binds the fer r i c ion more tightly than lactoferrin and EDDA, more weakly. The toxic effects of EDDA and HBED on animals have been studied. S t i f e l and Vetter (1967) drenched lambs with about 50 mg Kg body weight-1 day-1 of EDDA. This dose caused anorexia, awkward gait and increased urination. Two of 6 lambs receiving EDDA died. Postmortem examinations revealed hepatic and pulmonary toxicity and inflammation of the abomasal and intestinal mucosa. A single injection of 200 mg EDDA Kg body weight - 1 was not toxic to rats (Hershko et a l . 1984a). EDDA is not well absorbed when administered orally to rats (Hershko et al 1984b). Poor absorption from the digestive tract would minimize systemic toxic effects. HBED is less toxic than EDDA. The LDso of HBED for rats was 800 mg Kg body weight-1 (Grady and Jacobs 1981). HBED is also poorly absorbed from oral doses given to rats (Hershko et a l . 1986b). Miles and Khimji (1975) used EDDA as an indicator for synthesis of sidero-phores by bacteria. They found that 0.1 mg ml - 1 EDDA completely inhibited 7 x 104 Klebsiella spp.. Bacteria capable of synthesizing siderophores were not 55 inhibited by EDDA however. No studies on the antibacterial properties of HBED Table 8 . Comparison of st a b i l i t y constants of HBED and EDDA chelates (L'Eplattenier et a l . 1967). HBED EDDA Ion L L O £ _ K ! 1M L Log K ^ M L 5 Log K Mg2+ 10.51 8.00 2.51 Ca2+ 9.29 7.20 2.09 Cu 2 + 21.38 23.90 -2.52 Zn 2 + 18.37 16.80 1.57 FejLi- 39.68 33^9 15.77 I K " M L = . . [ . H L . 3 . THTTLT where [ML] = the concentration of the metal-ligand complex [M] = the concentration of the free metal [ L ] = the concentration of the free ligand have been published. These two iron chelators are promising candidates as replacements for lactoferrin in sow milk replacers. Vitamin B12 Binding Protein Like iron, vitamin B12 is tightly bound by a specific milk protein (Gregory and Holdsworth 1955). Sow's milk is particularly rich in B12 binding protein. Its B12 binding capacity i s 245 ng mL-1 compared to 80 ng mL-1 for human milk and 0.5 ng mL-1 for cow's milk. The distribution of B12 binding protein in the body is similar to that of slgA and lactoferrin.. It occurs in tears, saliva and gastric juice (Reiter 1985a). In vitro studies have shown that B12 binding protein is able to inhibit the uptake of vitamin B12 by a variety of bacteria including E. c o l i (Ford 1974). The antibacterial properties of B12 binding protein have not been studied. B12 binding protein also increases the piglet's absorption of vitamin B12 from the diet particularly in the f i r s t the f i r s t two weeks of l i f e (Trugo et a l . 1985). During this period the piglet does not synthesize adequate intrinsic factor and cannot therefore absorb vitamin B12. The piglet can absorb the vitamin bound to B12 binding protein however. Bi2 binding protein deserves further i n v e s -tigation in light of i t s relatively high concentration in sow's milk and i t s unknown antibacterial properties. Lvsozyme Lysozyme is an enzyme found widely in external secretions such as t e a r s , saliva, gastric secretions and milk. The concentration of lysozyme in the milks of different species varies considerably. In human milk, i t occurs at a l e v e l of 400 mg mL-1 while in bovine milk i t occurs at a level of 180 ug mL-1 (Jenness 1981). Lysozyme has several biological functions. The f i r s t involves i t s enzymic activity. Lysozyme hydrolyzes the 1-4 6 linkage between N-acetylgluc-osamine and N-acetylmuramic acid in the peptidoglycan layer of bacteria. This leads to lysis of the bacteria and prevents the outgrowth of spores and vegetative cells (Wasserfall and Teuber 1979). Lysozyme's ab i l i t y to prevent spore outgrowth may account for the low numbers or complete absence of C l o s t r i d i a in the feces of breast fed infants. The products of the hydrolysis of peptidoglycan by lysozyme also serve a protective function. The a c t i v e component in Freund's adjuvant was found to be N-acetylmuramyl-L-alanyl-D-isoglutamine (Genco et a l . 1983). This compound is similar to the products of the hydrolysis of peptidoglycan by lysozyme. Adjuvants stimulate antibody 57 responses and activate macrophages. Studies have shown that the products of hydrolysis of bacterial c e l l walls by lysozyme also have an adjuvant effect. The level of slgA in the feces of infants increased when lysozyme was included in the diet (Lodinova and Jouja 1977). Feeding the products of lysozyme hydrolysis of bacterial c e l l walls to mice resulted in increased levels of salivary slgA in mice (Morisaki et a l . 1983). IgM and complement in milk interact with lysozyme to increase i t s enzymic activity (Reiter 1985b). Gram negative bacteria have an outer membrane which can prevent lysozyme from penetrating to the peptidoglycan layer. IgM and complement produce lesions through the outer membrane of gram negative bacteria thus allowing lysozyme to attack i t s substrate (Reiter 1985b). Although lysozyme wi l l lyse bacteria in vitro there is s t i l l no evidence that this occurs in vivo. Its low concentration in sow's milk and colostrum indicates that i t is probably of l i t t l e importance as a protective mechanism in naturally suckled piglets. It may however interact with other components in milk and increase the overall level of protection. Its enzymic properties and the immunostimulatory effects of i t s hydrolysis products deserve further i n -vestigation. The Lactoperoxidase System The lactoperoxidase system has not been studied in sow's milk so the f o l -lowing discussion refers to work done with bovine and human milks. The lacto-peroxidase system consists of the enzyme lactoperoxidase, H2O2 and SCN-- Lacto-peroxidase and H2O2 form a complex which catalyzes the oxidation of SCN- to CO,,, + 2-NH^  , and S0 4 . The intermediary products of this reaction are inhibitory to bacteria. One of these products is OSCN- (Thomas et a l . 1981). It has a chaotropic effect on the inner membrane of bacteria (Reiter et a l . 1976). Gram 58 negative, catalase positive organisms such as E. c o l i require an exogenous source of H2O2 to be inhibited or k i l l e d by the lactoperoxidase system (Carlsson 1980) . Gram positive, catalase negative organisms excrete sufficient H2O2 to be self inhibitory under aerobic conditions (Reiter 1985b). An in vivo t r i a l with calves was conducted by Reiter and Harnulv (1982) . Calves were fed raw milk or milk heat treated to destroy the lactoperoxidase system. When the calves were challenged with oral E. c o l i , the calves that received the heat treated milk showed no reduction in bacterial numbers while the calves receiving raw milk had a 95% reduction in bacterial numbers. The addition of a magnesium peroxide to the diet as a solid source of H2O2 resulted in a reduction of 99.99% in bacterial numbers. Several other in vivo studies (Waterhouse and Mullen 1980) have shown that supplementing the natural lacto-peroxidase system in bovine milk by adding a source of H2O2 and SCN- decreased diarrhea and increased weight gains in calves. A milk replacer for calves con-taining the activated lactoperoxidase system is available in Sweden (Reiter 1985b). Experimental work on lactoperoxidase supplemented milk replacers for piglets is practical and desirable. Glycoconiuqate and Oligosaccharide Receptor Analogues Many pathogenic bacteria and bacterial toxins attach to the intestinal epithelium via glycoprotein, glycolipid and oligosaccharide receptors. (Holmgren et a l . 1983) . Analogues of these receptors are also present in milk. In vitro studies have shown that these molecules w i l l inhibit hemagglutination of red blood cells by E^ . c o l i (Holmgren et a l . 1983). Hemagglutination of red blood cells mimics adherence of bacteria to intestinal c e l l walls. In addition, the E. c o l i K88 adhesin which is specific for pigs was found to bind specifically to a receptor analogue on fat globule membranes from sow's milk but not cow or human milk (Reiter 1985b). These receptor analogues may protect suckling animals from bacterial and v i r a l infections by inhibiting attachment to the intestinal epithelium. In vivo studies on the receptor analogues have yet to be done. Sow's milk and colostrum contain a large number of factors that protect the nursing piglet from bacterial and v i r a l infection. These factors interact extensively and i t is probable that the protective value of the whole is greater than the sum of the individual factors alone. The practical application of our knowledge of these protective systems does not require that they a l l be replaced so that a milk replacer is a perfect copy of maternal milk. Porcine serum immunoglobulins and bovine colostrum have been successfully used as immunoglobulin sources for colostrum deprived piglets. Bovine serum immunoglobulins have not been investigated however. A direct comparison of bovine and porcine serum immunoglobulins would show whether bovine serum is a large potential source of immunoglobulins for sow milk replacers. The synthetic iron chelators EDDA and HBED have an aff i n i t y for iron similar to that of lactoferrin. They may provide an inexpensive way of mimicking the antibacterial properties of lactoferrin in sow milk replacers. Viable cells in sow's milk may enhance the piglet's immune system. While the cells in sow's milk are a selected, specialized population, peripheral blood leukocytes may provide a suitable replacement just as serum immunoglobulins can replace colostral im-munoglobulins. The following experiments w i l l investigate i) bovine versus porcine serum immunoglobulins, i i ) the synthetic iron chelators EDDA and HBED as replacements for lactoferrin and i i i ) peripheral blood leukocytes as replacements for co-lostral leukocytes. 60 EXPERIMENT 1 Introduction Immunoglobulins from many sources have been used to provide passive immunity to colostrum deprived piglets with varying results. Porcine immunoglobulins from unfractionated serum provides adequate immune protection but i t s feeding i s impractical due to palatability problems and relatively low immunoglobulin content (Scoot et a l . 1972). Ammonium sulphate fractionation of immunoglobulin from serum provides a cheap simple method of concentrating serum immunoglobulin. Immunoglobulins from bovine sources have also been used as sources of passive immunity for piglets. There i s interest in bovine immunoglobulins for several reasons:(1) the availability of products like bovine plasma, colostrum and acid whey; and 2) the possible use of single immunoglobulin source for both piglet and calf milk replacers. The objective of Experiment 1 was to compare immunoglobulins isolated from bovine and porcine serum as sources of passive immunity to colostrum deprived piglets. The immunoglobulins used in this experiment were prepared using polyphosphate fractionation. This was the f i r s t time immunoglobulins prepared by this method were used as a source of passive immunity for piglets. Materials and Methods  Experimental Design There were 3 treatments in which piglets received diets as shown in Table 9 with 3 outcome groups of 11,11 and 12 piglets based on the availability of l i t t e r s . The term outcome group refers to a group of pigs that started the experiment at the same time. 6 1 Table 9 . Experimental protocol for studying the administration of immunoglobulins to piglets (Experiment 1). Treatment Age of Piglets (d)  Name 1 2 to 14 15 to 28 Control No Ig No Ig No Ig Bovine 120 BIgG 4 BIgG No Ig Porcine 20 PIqG 4 mg PlaG No Ig 1 A l l values are in mg mL-1 These levels were based on results by McCallum (1977) who found that 26.7 mg mL-1 of PIgG on day 1 and 5.33 mg mL-1 on days 2-20 provided adequate protection for colostrum deprived piglets. Levels of 13.3 mg mL-1 on day 1 and 2.7 mg mL-1 on days 2-20 provided inadequate protection. The levels in this study were set between these two values to see i f an intermediate value would provide adequate immunity. Results were analyzed using the General Linear Models procedure of SAS (Statistical Analysis System Institute Inc. 1985). The following least squares model was used to analyze the data in Experiment 1. Yu = u + Ti + Gj + TiGj + Eij where Yu = dependent variable u = overall mean Ti = effect of the ith treatment Gj = effect of the jth outcome group 62 TiGj = interaction of the i t h treatment with the jth outcome group Eij = residual error for each sample The interaction TiGj was not significant for any of the measures so i t was added to the error term to increase precision and the results were recalculated. Analysis of covariance was performed on average daily gains using i n i t i a l weight as covariate. Survival was analyzed by assigning 1 to piglets that survived the experiment and 0 to those that died. Differences between treatment means were analyzed using orthogonal contrasts. Preparation of Porcine and Bovine Immunoglobulins Citrated porcine whole blood and bovine plasma were obtained from Inter-continental Packers Ltd. of Vancouver, B.C.. Throughout the entire procedure a l l materials were kept at 4 °C to minimize bacterial growth. The porcine blood was allowed to settle overnight and the plasma was siphoned off the top. Fibrin was precipitated by the addition of calcium carbonate and removed by centri-fugation. Sera were fractionated using sodium polyphosphate "glass", a mixture of NaiaPiaCMo-NaaoPiaOog (Lee et a l . 1988). One hundred mL of a solution con-taining 114.4 g L _ 1 of sodium polyphosphate glass (Sigma Chemical Col, St.louis, MO) and 84.85 g L _ 1 NaCl were added to each L of plasma with constant s t i r r i n g . After pH adjustment to 3.95 with 3M HCl, the mixture was stirred for 10 minutes and allowed to equilibrate overnight. The mixture was then centrifuged at 3000 g for 5 minutes and the supernatant retained. The immunoglobulin in the super-natant was concentrated using a Pellicon u l t r a f i l t r a t i o n system (Millipore Corporation, Bedford, MA) with a 100,000 nominal molecular weight limit f i l t e r . The concentrated fraction was lyophilized and stored at -18 °C in sealed plastic containers. 63 Assay of Administered and Serum Immunoglobulins Quantitative analyses of bovine IgG and porcine IgG were done using the double antibody sandwich enzyme linked immunosorbent assay (ELISA) described by Voller et a l . (1976). Antibodies were obtained from Sigma Chemical Co., St. Louis MO.. Antibodies to bovine IgG w i l l also cross react with porcine IgG. For the bovine IgG assay, the cross reacting antibodies were removed by passing anti-bovine IgG anti-bodies over a column containing porcine serum proteins cross linked to agarose. Any cross reacting antibodies are bound in the column. Likewise for the porcine IgG assay anti-porcine IgG antibodies were passed over a column containing bovine serum proteins cross linked to agarose. IgA and IgM were not assayed. Animal Management Piglets were removed from sows immediately after birth and placed in boxes. A l l piglets over 800 g with no obvious anatomical or physiological defects (splay legs, atresia anii etc.) were used. When sufficient piglets were born for the outcome group they were randomly placed in individual cages measuring 46 x 92 cm. The treatments were then assigned to the cages using a random number generator. The cages allowed contact between adjacently housed piglets. The room containing the cages was maintained at 32 °C. The cages were sanitized using high pressure washing and a quaternary ammonia disinfectant before the piglets were placed in them. Milk Replacer The basal diet was a commercial, non-medicated, a l l milk protein sow milk replacer (Van Waters and Rogers Ltd., Vancouver B.C.). The milk replacer con-tained 21% crude protein, 31% fat, and NRC requirement levels of a l l vitamins and minerals except iron. For this experiment 55 mg of iron as FeCl3 was added 64 for each Kg of milk replacer. The exact formulation of the diet i s proprietary information. A commercial milk replacer was used because this i s the type of diet that would be used in any commercial applications of immunoglobulin for-t i f i e d milk replacer. Results obtained with a purified laboratory diet might be less applicable to commercial applications. The dry diet was mixed with water to provide a 20% solids milk replacer. Freeze dried immunoglobulins were mixed with the dry milk replacer to provide the appropriate concentration. Feeding Regimen Piglets were nipple fed by hand for the f i r s t 72 hours. They were fed hourly for the f i r s t 6 hours and thereafter every 3 hours. After the period of hand feeding, the piglets were fed from a nursing bottle located in the cage and f i l l e d hourly by a peristaltic pump up to day 14. From day 15 to 28 they were fed unfortified milk replacer twice a day from piglet waterers equipped with metal nipples. On day 1 piglets were fed ad l i b . On days 2 to 28 they were fed 375 mL Kg - 1 body weight day - 1 to a maximum of 1,500 mL day - 1. The piglets were weighed every other day and the feed intake was adjusted accordingly. This level of intake i s the same used by McCallum (1977) and provides less feed than naturally reared piglets receive from a sow. Diarrhea Scores Diarrhea was estimated using a scale described by Nocek et a l . (1984): 1) normal, no f l u i d 2) soft, mostly solid 3) runny, mostly f l u i d 4) watery, a l l f l u i d 5) watery, blood A l l observations were done by the same observer. 65 Protein Analyses Protein was determined using the macro Kjeldahl method. Ash Content The ash content of the bovine and porcine immunoglobulin concentrates was measured by heating the samples at 550 °C and determining the resultant change in weight. Blood Samples Blood samples were taken from the anterior vena cava at birth and on days 1, 4, 7, 14, 21 and 28. The blood was collected in 100 uL heparinized capillary tubes. The tubes were sealed with Critoseal and centrifuged for 15 minutes in a Canlab International Hicrocapillary Centrifuge (Model MB) and read on a Canlab reader (Model (CR). The plasma samples were then stored at -18 °C. Post-Mortem Examinations Post-mortem examinations were performed within 24 hours of death at the Provincial Veterinary Pathology Laboratory (Abbotsford, B.C.) on a l l piglets that died during the experiment. Bacterial cultures of organisms found in the intestine, lung, blood, spleen and kidneys were taken routinely. . Results Bovine and Porcine Immunoglobulins The yield of freeze dried product was about 800-900 g for 120 L of porcine blood and 3100 g for 120 L of bovine plasma. The analysis of the freeze dried product i s shown in Table 10. Gel electrophoresis was performed on porcine and bovine serum, and on both the precipitate and supernatant fractions from the polyphosphate fractionation (see Figures 11 and 12). Survival Only 2 of 11 piglets in the control group survived the entire 28 day period 66 (see Table 11). The mean survival for the control group was significantly lower than for the bovine or porcine immunoglobulin treatments. The difference in sur-vival between the porcine and bovine treatments was not significant. There was a trend to higher survival in the porcine group however. The apparent cause of death in the f i r s t and second outcome groups was E. c o l i septicemia. Of the 5 piglets that died in the third outcome group only two died of E± c o l i sep-ticemia. Both of these were controls. The other control piglet that died in this outcome group died of salmonellosis. One bovine treatment piglet died in the third outcome group. The pathogens isolated from i t s tissues were E;. c o l i , Klebsiella spp., hemolytic Staphylococcus spp. and Pseudomonas spp.. The cause of death of the porcine treatment piglet that died was Streptococcus suis type II. Diarrhea The piglets fed porcine or bovine immunoglobulins did not differ sig-nificantly during any period (see Table 12). The controls had significantly more diarrhea during weeks 1 and 2 than the bovine or porcine immunoglobulin fed groups. Average Daily Gains and Piglet Weights Average daily gains and piglet weights were both analyzed. This was done because groups with small but non-significant differences in average daily gains can differ significantly in weight at the end of the experiment. The controls had significantly lower average daily gains during weeks 1,2 and 3 than the bovine and porcine immunoglobulin fed piglets (see Table 13). The bovine immunoglobulin fed piglets had significantly lower average daily gain than those that received porcine immunoglobulins during week 1. There were no treatment differences for birth weights (see Table 14). The 67 controls had significantly lower mean weights on a l l days and had a mean weight of only 3,130 g at 28 days of age. The bovine immunoglobulin fed piglets had significantly lower weights than the porcine immunoglobulin fed piglets on day 7. On days 14, 21 and 28, however, there were no significant differences among outcome groups. Plasma Immunoglobulins Controls had no measurable plasma PIgG until 7 days of age (2.3 mg mL-1) (see Table 15 and Figure 13). This rose rapidly to 15.3 mg mL-1 on day 21 and 16.4 mg mL-1 on day 28. The porcine immunoglobulin fed group showed a rapid rise in plasma PIgG levels to 19.7 mg mL-1 at day 1. The level dropped to a minimum value of 8.3 mg ml - 1 on day 21. The controls had significantly higher plasma PIgG levels on days 21 and 28 than the porcine group. The plasma of the bovine immunoglobulin fed piglets was analyzed for both BIgG and PIgG. On day 1, the plasma levels of BIgG of these piglets were sig-nificantly less than PIgG levels in the porcine immunoglobulin fed piglets. The level of PIgG in the bovine immunoglobulin treatment piglets was not measurable until day 4. It rose quickly to a mean of 13.1 mg ml - 1 on day 21. The plasma PIgG of the bovine immunoglobulin fed group was not significantly different from the other groups on days 21 and 28. The level of BIgG had declined to 0.2 mg mL-1 by day 28. Discussion Survival of the control group is an indication of the level of environmental pathogens. Varley et a l . (1986) achieved survival rates of 55-98% for immunoglobulin deprived piglets in extremely clean environmental conditions. The pigs were reared in isolated cages with f i l t e r e d a i r . Under conditions more 68 typical of commercial units, McCallum et a l . (1977) obtained survival rates of 0-15% . The survival rate of 22% in the present experiment suggests a high level of environmental pathogens. A negative control group has the disadvantage of acting as a reservoir of contamination. Fecal cross contamination presumably allowed the spread of large numbers of pathogenic organisms from the controls to adjacent piglets. When McCallum et a l . (1977) deleted controls from one t r i a l , the survival rates and average daily gains of the piglets increased markedly. There were no significant differences in survival between the three outcome groups but the cause of death changed as the experiment progressed. The apparent cause of death for piglets in the f i r s t outcome group was Ej. c o l i infection. Piglets in the second replicate died at an earlier age but the agent remained E. c o l i . In the third outcome group deaths were at about the same age as in the second. The organisms responsible included Streptococcus suis, Klebsiella spp., Pseudomonas spp., hemolytic Staphylococcus spp. and Salmonella spp.. This pattern suggests f i r s t a build up of numbers of Ej. c o l i and then an increase in other less common environmental pathogens. An a l l i n - a l l out regimen with thorough sanitizing would probably reduce the build up of disease in the experimental room. Sampling blood from the vena cava i s stressful to the piglets and stress can decrease disease resistance. However, this method of blood sampling probably did not contribute to the low survival rate of the control piglets. In a similar type of experiment (Scoot et a l . 1972) where no blood samples were taken, control piglets had a survival rate of 0%. Diarrhea scores for the control group averaged 3.1 during the f i r s t week of l i f e . Control piglets a l l had diarrhea scores of 4 before death. In contrast 69 the bovine immunoglobulin fed piglets that died had diarrhea scores of only 2 or 3. The mean score for this group was the same as for the porcine immuno-globulin fed piglets. This indicates that although BIgG provides poor systemic protection i t provides adequate local protection in the lumen of the small intestine. The growth rates of the bovine and porcine fed piglets differed significantly only during the f i r s t week. On weeks 2, 3 and 4 the average daily gains of the bovine immunoglobulin fed group were not significantly different than the group that received porcine immunoglobulins. There was no compensatory gain and the weights of the bovine immunoglobulin fed group stayed about 400 g less than those that received porcine immunoglobulins from day 7 to 28. The growth rates of piglets in this t r i a l were similar to those reported by McCallum et a l . (1977) but lower than those for naturally reared piglets. Increasing the level of intake would probably improve the gains. Braude et a l . , (1970) obtained much higher average daily gains with higher intakes. They also noted higher mortality and incidence of diarrhea at these levels. The plasma PIgG levels of piglets that died on the control and porcine immunoglobulin treatments were markedly lower than those of survivors. The one porcine immunoglobulin fed piglet that died had a plasma PIgG level of only 9.0 mg mL-1 on day 1. This was the lowest level of any piglet on this treatment. The two surviving controls were the only ones to have measurable plasma PIgG by day 7 ( 4.1 and 2.6 mg mL - 1). A l l other controls were either dead or had plasma PIgG levels less than 0.1 mg mL-1 at 7 days of age. There was no apparent relationship between the total plasma IgG and survival in the bovine group. The plasma PIgG levels of the porcine immunoglobulin fed group were lower than published reports for naturally reared piglets. Klobassa et a l . (1981) 70 measured PIgG levels in naturally suckled piglets and found that they peaked at 40.2 mg mL-1. This is more than double the 20.8 mg mL-1 in the present study. This i s because sow colostrum provided a much higher level of immunoglobulins than in this experiment. The average IgG concentration of sow's colostrum immediately following parturition was found to be 95.6 mg mL-1 in one study Klobassa et a l . 1091). This decreased to 14.2 mg mL-1 at 24 hours after parturition. While i t is possible to increase the level of IgG during the day 1, there may be no advantage in doing so. In fact, there may be a disadvantage. Henry and Jerne (1968) found that passively administered IgG inhibits the active synthesis of IgG and IgM of the same specificity. In the present study, less IgG may have provided less inhibition to the development of active immunity by the piglet. The naturally reared piglets studied by Klobassa et a l . (1981) had a plasma IgG level of only 6.7 mg ml - 1 on day 28. This did not increase until after 5 weeks. The plasma IgG level of the porcine group began increasing after 3 weeks of age and reached 9.5 mg mL-1 on day 28. The optimum level of pass-ively administered immunoglobulin would give adequate protection from disease and minimize inhibition of the development of active immunity. A l l PIgG in the plasma of the bovine immunoglobulin fed piglets was syn-thesized by the piglet. Measurable levels of PIgG occurred by 7 days of age. It is possible to approximate the PIgG synthesized by the porcine fed piglets using the half l i f e of 14 days for PIgG in the bloodstream (Curtis and Bourne 1973). The calculated values show that no significant synthesis took place until after 21 days of age. Precocious development of active immunity by the bovine immunoglobulin fed group could be due to the low i n i t i a l level of plasma BIgG. It could also be that BIgG is not as inhibitory to active immunoglobulin synthesis. Henry and Jerne (1968), found that rabbit IgG inhibited synthesis 71 of Ig of the same specificity in mice. Cross-species inhibition may also take place between bovine and porcine IgG. The confounding effect of the low levels of BIgG in the bovine group makes this impossible to determine in the present study. The synthesis of PIgG by the control piglets was even more rapid than by the bovine immunoglobulin fed group. The difference was not significant however. In this case there was no passively administered immunoglobulin so there was no inhibition to the development of active immunity. The means for the controls on days 14,21 and 28, are based on only two piglets however. The absorption of BIgG by the bovine immunoglobulin fed group was much lower compared to PIgG absorption by the porcine immunoglobulin fed group. Leary and Lecce (1979) found that the piglet selectively absorbs PIgG when other macromolecules are present in the lumen of the intestine. Their proposed mech-anism involves the presence of specific PIgG receptors on the surface of enter-ocytes which bind PIgG selectively. The non-immunoglobulin macromolecules stimulate pinocytosis but they are not bound by specific receptors and hence not absorbed as effi c i e n t l y . The low absorption of BIgG in this study may have been due to the in a b i l i t y of these receptors to bind and transport BIgG. In conclusion, bovine serum immunoglobulins are not well absorbed from the diet when fed to colostrum deprived piglets during the f i r s t day of l i f e . Bovine serum immunoglobulins cannot therefore replace porcine serum immunoglobulins on day 1. The function of dietary immunoglobulins after day 1 does not depend on absorption however. The similar levels of diarrhea seen on days 2-14 in both bovine and porcine immunoglobulin fed piglets may indicate that either source of immunoglobulin w i l l provide adequate immune protection in the small intestine after the f i r s t day of l i f e . Table 10. Composition of freeze dried bovine and porcine immunoglobulin concentrates 1. Component Bovine Iq Porcine Iq Crude Protein 2 (%) 74.5 77.9 IgG (%) 53.5 37.0 Ash (%) 12.4 10.1 Iron (uq/q) 29.8 17.6 1 Analyses were performed on pooled samples prepared for Experiment 1. 2 Analyses are on a dry matter basis. Table 11. The effect of bovine or porcine iinroioglobulins on the survival of colostrum deprived piglets (Experiment 1). Number Surviving at day Birth 7 14 21 28 Treatment N N N N N % SE Control 11 5 3 2 2a 19.4 11.9 Bovine Ig 11 10 8 8 8 72.2 11.9 Porcine Ig 12 12 11 11 11 91.6 11.3 N = the number of surviving piglets. a The contrast Control vs Bovine Ig and Porcine Ig is significant (P < 0.05). Table 12. The effect of bovine or porcine imnuncglobulins on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 1). week 1 2 3 4 Treatment X SE X SE X SE X SE Control 3.1a 0.05 2.4a 0.16 1.5 0.15 1.1 0.11 Bovine Ig 1.0 0.10 1.1 0.08 1.2 0.07 1.2 0.06 Porcine Ig 1.1 0.04 1.1 0.07 1.1 0.06 1.0 0.00 a The contrast Control vs Bovine Ig and Porcine Ig is significant (P < 0.05). 74 Table 13. The effect of bovine or porcine immunoglofcilins on the average daily gains of colostrum deprived piglets (Experiment 1). Average Daily Gains (g dayl) • week 1 week 2 week 3 week 4 Treatment X SE X SE X SE X SE Control Oa 12 50a 33 60a 43 153 24 Bovine Ig 90b 24 150 16 190 21 200 12 Porcine Ig 150 9 140 12 200 17 190 9 a The contrast Control vs Bovine Ig and Porcine Ig is significant (P < 0.05). b The contrast Bovine Ig vs Porcine Ig is significant (P < 0.05). Table 14. The effect of bovine or porcine imminoglobulins on the body weights of colostrum deprived piglets (Experiment 1). Body Weights (g) Birth day 7 day 14 day 21 day 28 Treatment X SE X SE X SE X SE X SE Control 1,290 58 1,280a 171 1,650a 342 2,060a 540 3,130 614 Bovine Ig 1,210 62 1,860b 81 2,900b 164 4,200b 258 5,580 293 Porcine Ig 1.240 55 2.320 66 3.290 132 4.670 208 5.980 236 a The contrast Control vs Bovine Ig and Porcine Ig is significant (P < 0.05). b The contrast Bovine Ig vs Porcine Ig is significant (P < 0.05). 75 Table 15. The effect of bovine or porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 1). Age of Piglets (days) 0 1 4 7 14 21 28 Treatment X SE X SE X SE X SE X SE X SE X SE Control PIgG <o.i - <0.1a - <0.1a - 2.3a 2.6 6.8 2.6 15.3a 3.7 16.4a . 3.7 Bovine PIgG <0.1 - <0.1 _ <0.1 _ 1.1 1.3 4.3 1.3 12.0 1.9 13.7 1.9 Bovine BIgG <0.1 - 4.0 0.5 5.9 0.4 5.5 0.5 2.7 0.4 0.4 0.2 0.2 0.1 'Bovine Total <o.i - 4.0b 0.5 5.9b 0.4 6.6b 1.3 7.0b 1.4 12.4 2.0 13.9 1.9 Porcine PIgG <0.1 - 20.8 1.1 18.8 0.9 14.1 1.1 11.0 1.1 8.3 1.6 9.5 1.6 'For the Bovine Ig treatment, the total of PIgG and BIgG was compared to the other treatment means. a The contrast Control vs Bovine Ig and Porcine Ig is significant (P < 0.05). b The contrast Bovine Ig vs Porcine Ig is significant (P < 0.05). Figure 11. Gel electrophoresis of polyphosphate fractions from porcine serum. (1) porcine serum; (2) polyphosphate supernatant fraction; (3) polyphosphate precipitate fraction. A = Albumen; Ig = Immunoglobulin. A Ig A Ig Figure 12. Gel electrophoresis of polyphosphate fractions from bovine serum. (1) bovine serum; (2) polyphosphate supernatant fraction; (3) polyphosphate precipitate fraction. A = Albumen; Ig = Immunoglobulin. A Ig 78 Figure 13. The effect of bovine or porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 1). Day 79 EXPERIMENT 2 Introduction Lactoferrin i s an iron-binding protein found in sow's milk and colostrum. It inhibits the growth of bacteria in the intestinal tract of the suckling piglet. Supplementation of milk replacers with lactoferrin may improve per-formance of piglets reared on a r t i f i c i a l diets. This use of lactoferrin may be unfeasible however. There are several problems with using lactoferrin as feed additive to piglet milk replacers. F i r s t l y , lactoferrin has a high molecular weight (93,000 daltons). This means that about 833 g of lactoferrin are required to bind 1 g of iron. Secondly, the iron content of most commercial milk replacers i s relatively high even i f no iron is added. Contamination of milk replacers with iron during processing boosts the level of iron many times that of sow's milk. These two factors taken together mean that a very large amount of lactoferrin would be required to bind the iron in commercial sow milk replacers. Lactoferrin i s uneconomical at the present time. Lactoferrin's anti-bacterial effect i s due only to i t s a b i l i t y to bind iron. Other synthetic iron chelators may be able to replace lactoferrin in a r t i f i c i a l diets. Two such compounds are ethylene diamine-di-orthohydroxyphenyl acetic acid (EDDA) and N,N'-Bis(o-hydroxybenzyl)-ethylenediamine diacetic acid (HBED). The objective of this experiment was to compare in vitro, the antibacterial effect of lactoferrin, EDDA and HBED, with and without added immunoglobulins. Materials and Methods  Experimental Design The treatments used are shown in Table 16. The results were analyzed 80 Table 16. Experimental protocol for testing the effect of lactoferrin, EDDA and HBED on the growth rate of E^ _ c o l i 0 157 K88 (Experiment 2). PIgG Chelator 0 mg/mL 4 mg/mL Control 0.87 ug/mL Fe A 0.94 mg/mL Fe 10.0 mg/mL lactoferrin 11.9% 12.9% 0.1 mg/mL EDDA 5.6% 6.0% 0.1 mg/mL HBED 6.1% 6.6% A The Control values represent the amount of iron in the growth medium. The values for lactoferrin, EDDA and HBED represent the percent saturation of the chelators with iron. using the General Linear Models procedure of SAS (Statistical Analysis System Institute Inc. 1985) using the following least squares model. Yijt = u + Ci + Pj + CiPj + Eij where Yijt = Log(CFU) mL"1 at time t u = the overall mean Ci = the effect of the i t h chelator Pj = the effect of the j t h level of PIgG CiPj = the interaction between the i t h chelator and the jth level of PIgG Ei j = the residual error for each sample The Logio of the colony-forming units mL-1 (Log(CFU) mL-1) was the dependent variable and represents the number of livin g bacteria or clumps of bacteria in 81 the growth medium. Treatment means were compared using the Ryan-Einot-Gabriel-Welsch multiple F test (Statistical Analysis System Institute Inc. 1985). Preparation of Lactoferrin Lactoferrin was separated from bovine colostrum using a method described by Law and Reiter (1977). Colostrum (7.5 L) was adjusted to pH 7.0 by the addition of K H C O 3. Ammonium ferrous sulphate (8.6 g) was added to saturate the lactofer-rin with iron. Dry CM-Sephadex (90 g) (C-50, Pharmacia Fine Chemicals AB, Uppsala, Sweden) prepared in gel form in 0.02 M potassium phosphate buffer (pH 7.0) was added to the colostrum and the mixture stirred for 2 hours. The CM-Sephadex was allowed to settle and the supernatant was decanted. This was followed by 3 washings of the Sephadex with d i s t i l l e d water. The Sephadex was then applied to a column at room temperature and eluted with 0.02M phosphate buffer containing 0.2M NaCl (pH 7.0) un t i l no protein was detected in the washings, based on absorbance at 280 nm. The Sephadex was then eluted with a 0.02M phosphate buffer containing 0.5M NaCl (pH 7.0) and the lactoferrin containing fraction collected. Iron was removed from the lactoferrin by dialysis against trisodium citrate overnight followed by dialysis against deionized water for 24 hours at 4 °C. The lactoferrin was then freeze dried and stored at -18 °C. The unsaturated iron binding capacity of the lactoferrin was measured using a method described by Bullen et a l . (1972). A 1% solution of lactoferrin was prepared in phosphate buffered saline (0.01M phosphate, 0.15M NaCl, pH 7.4). The absorbance of the lactoferrin at 470 nm was measured during titration with 1.0 mM ferric nitrilotriacetate (pH 7.4). The unsaturated iron binding capacity was calculated from a plot of absorbance vs the amount of iron added. The unsaturated iron binding capacity was found to be 1 g iron per 1370 g of lac-82 toferrin. The theoretical value for the unsaturated iron binding capacity for iron free lactoferrin, using a molecular weight of 93,000 daltons (Weiner and Szuchet 1975), i s 1 g iron per 833 g of lactoferrin. Synthetic Iron Chelators EDDA was purchased from the Sigma Chemical Co. (St Louis, MO) in 90% pure form and purified using the method of Rogers (1973). HBED was obtained as a gif t from the Strem Chemical Co (Newbury, MA) in a pure iron free form. Both chelators were assayed for iron content prior to the experiment. Bacteria E. c o l i 0 157 K88 was obtained from the British Columbia Veterinary Path-ology Laboratory. It i s a known pathogen for pigs. Cultures were stored on Trypticase Soy Agar (Difco Laboratories Inc., Detroit MI) slants at 4 °C and transferred monthly. Inhibition Assays Trypticase Soy broth, lactoferrin, EDDA, HBED and immunoglobulins were sterilized by passage through a 0.22 um membrane f i l t e r (Millipore Corporation, Bedford MA).The iron content of the trypticase soy broth and the additives was measured. The liquid trypticase soy broth contained 0.87 ug mL-1 of iron. The PIgG contained 0.07 mg mL-1 of iron at the concentration i t was used. Hemoglobin was not present at a detectable level in the PIgG solution. None of the other additives contained detectable levels of iron. E;. c o l i 0 157 K88 were grown overnight in Trypticase Soy broth at 37 °C. The culture was then diluted to about 5 x 10s colony-forming units (CFU) mL-1. One mL of this culture was added to duplicate tubes containing 4 mL of trypticase soy broth and appropriate amounts of iron chelator and immunoglobulins for each treatment. The tubes were then incubated at 37 °C. Counts of viable bacteria 83 were done at 0, 2, 4, 6 and 12 hours on trypticase soy agar plates using a Spiral Plater (Model CU, Spiral System Instruments Inc., Bethesda, MD.). The plates were incubated 24 hours at 37 °C before counting. Iron Analyses Samples were ashed at 500 °C and dissolved in 4M HC1. The sample was then analyzed using an atomic absorption spectrophotometer. Results There were no significant treatment differences in the Log(CFU) mL-1 at 0 hours (see Table 17 and Figure 14). The overall mean Log(CFU) mL-1 was 5.01 (1.02 x 109 CFU mir 1). The PIgG treatment had significantly fewer Log(CFU) ml - 1 than the control treatment at 2,4 and 6 hours. At 12 hours there was no significant difference between the two treatments. The HBED treatment was not significantly different from the control treatment at any sampling time. Likewise, the HBED-PIgG treat-ment was not significantly different from the PIgG treatment. The lactoferrin treatment was not significantly different from the control treatment at 2 hour incubation. At 4, 6 and 12 hours incubation however i t had significantly fewer Log(CFU) ml - 1 than the control treatment. The lactoferrin treatment had significantly more Log(CFU) ml - 1 at 2 hours than the PIgG treat-ment. At 4 and 6 hours incubation, there were no significant differences between the two treatments. At 12 hours the lactoferrin treatment had significantly less Log(CFU) ml - 1 than the PIgG treatment. The lactoferrin-PIgG and the lactoferrin treatments had significantly different Log(CFU) ml"1 only after 6 hours incubation. At that time lactoferrin-PIgG had fewer Log(CFU) ml - 1 than the lactoferrin treatment. Lactoferrin-PIgG also had significantly fewer Log(CFU) ml"1 than the PIgG treatment after 6 84 hours. This may demonstrate synergism between lactoferrin and PIgG. The results for EDDA were similar to the results for lactoferrin. The EDDA treatment had significantly fewer Log(CFU) ml - 1 than the control treatment at 4, 6 and 12 hours. The EDDA treatment also had significantly fewer Log(CFU) ml - 1 than the PIgG treatment at 6 and 12 hours. The EDDA and the EDDA-PlgG treatments were significantly different only after 4 and 6 hours incubation. At that time the EDDA-PIgG treatment had significantly fewer Log(CFU) ml - 1 than the EDDA treatment. The EDDA-PIgG treatment also had sig-nificantly fewer Log(CFU) ml - 1 after 4 and 6 hours than the PIgG treatment. Once again this may demonstrate synergism at these times. Discussion Sows milk is bacteriostatic or bacteriocidal for many porcine strains of E. co l i (Nagy et a l . 1976a). Immunoglobulins and lactoferrin are major factors in this effect. There i s disagreement whether either alone is sufficient to inhibit bacterial growth or i f inhibition requires both immunoglobulins and lactoferrin. Wilson (1972) found that immunoglobulins alone are bacteriostatic. Whey from bovine colostrum or milk caused bacteriostasis of an E^ c o l i strain after cows were immunized for that strain. In another study the bactericidal effect of bovine colostral whey was inactivated by heating to 56 °C (Reiter and Brock 1975). This treatment did not affect immunoglobulins but did destroy complement. Immunoglobulins with no complement caused a temporary inhibition of the growth of Ej. c o l i similar to that in the present experiment. Other studies found that immunoglobulins alone did not affect bacterial growth at a l l (Spik et a l . 1978, Stephens et a l . 1980). Rainard (1986a) compared the effects of bovine IgGi and lactoferrin on the growth of E. c o l i and found bovine IgGi alone was bacteriostatic. They then used 85 a sonic treatment that broke up chains and clusters of bacteria before per-forming a viable count. Under these conditions, the immunoglobulin had no effect on bacterial growth. Rainard suggested that the effect of immunoglobulins is due to the formation of microcolonies by agglutination rather than true inhibition of growth. The bacteria are s t i l l able to multiply however. As the density of immunoglobulins on the surface of bacteria decreases the micro-colonies break up and the apparent inhibition ends. Microscopic examination of E_j. c o l i grown with PIgG in the present experiment did reveal clusters of bacteria after 6 hours incubation. While immunoglobulins may have no effect on actual numbers of bacteria, the formation of microcolonies of bacteria may be of benefit to the suckling piglet. Immunoglobulins coat the microcolonies of bacteria. This could prevent bacteria from adhering to the small intestine and causing disease (Nagy et a l . 1976b). The immunoglobulins might protect the piglet until the microcolonies break up. Apparent inhibition of growth would show how long the immunoglobulins coat the bacteria and protect the piglet. In the present experiment, immunoglobulins slowed apparent bacterial growth for 6 hours. At 12 hours the Log (CFU) ml - 1 was the same as for the control treatment. Studies of lactoferrin also show conflicting results. Some experiments showed that lactoferrin is more effective when immunoglobulins are also present (Rogers 1973; 1976, Spik et a l . 1978). Other studies show no additive effect between immunoglobulins and lactoferrin. (Reiter et a l . 1975, Samson et a l . 1979, Rainard, 1986a). Interestingly, Spik et a l . (1978) found that lactofer-r i n was more inhibitory in milk deactivated at 100 °C than in peptone water. They speculated that milk either contributes directly to lactoferrin's activity or indirectly stabilizes lactoferrin's structure. 86 The conditions of the experiment have much to do with the results obtained. The present study found an additive effect between immunoglobulins and lactoferrin only at 6 hours. Lactoferrin had no significant effect on bacterial growth during the f i r s t 2 hours. Mellencamp et a l . (1981) found bacteria can use stored iron for growth even in the presence of lactoferrin. Bacteria grown in iron poor conditions lose this internal iron store. Lactoferrin can then cause immediate and complete bacteriostasis. Rainard (1986a) grew bacteria in an iron poor media. The effect of lactoferrin alone was so great that i t may not have been possible to detect any added effect of immunoglobulins. In an iron rich environment lactoferrin i s only effective after the iron stores are depleted. Since immunoglobulins are effective for only a limited period, the length of time an additive effect can occur i s small. The sampling time and level of iron in the environment determine whether an additive effect occurs. EDDA and lactoferrin inhibited growth during the 6 to 12 hour period while PIgG did not. Once a l l of the available immunoglobulin was bound to bacteria, i t was no longer effective at preventing further bacterial growth. Iron chelators have a longer lasting effect on bacterial growth. This has sig-nificance for the feeding of immunoglobulin f o r t i f i e d milk replacers to animals. The interval between feedings must be short to be sure of a steady supply of unbound immunoglobulins. EDDA caused complete bacteriostasis over a 12 hour period. As mentioned in the literature review, Miles and Khimji (1975) used EDDA as an indicator for the synthesis of siderophores by bacteria. They found that 0.1 mg ml - 1 EDDA completely inhibited 7 x 104 Klebsiella spp.. Bacteria capable of synthesizing siderophores were not inhibited by EDDA. A synergistic effect between EDDA and 87 PIgG was apparent in the present study at 4 and 6 hours. The effect was not apparent at 12 hours but by this time supply of available PIgG may have been exhausted. EDDA was able to completely inhibit the growth of the test organism with or without PIgG. Lactoferrin was not as effective in inhibiting the growth of the test organism. EDDA is a small molecule compared to lactoferrin. It may be able to diffuse into the bacterial c e l l affecting iron metabolism within the c e l l . This may account for i t s superior inhibition of bacterial growth compared to lactoferrin. HBED had no effect on bacterial growth. Since i t has a stronger af f i n i t y for iron than either lactoferrin or EDDA this i s surprising. No information on why HBED was ineffective at preventing bacterial growth was available in the literature or from the manufacturer. From these results i t seems clear that EDDA can mimic lactoferrin's anti-bacterial effect in vitro for this strain of E^ c o l i . HBED did not have any effect at a l l on bacterial growth. Depending on the performance of EDDA in vivo, i t may be an effective antibacterial agent in the diets of a r t i f i c i a l l y reared piglets. 88 Table 17. The effect of lactoferrin, EDDA and HBED, with and without porcine immunoglobulins, on the Mean Log(CFU) mL"1 of EL coli 0 157 K88 (Experiment 2). Treatment Chelator PIgG Time : (Hours) (mg/ml) (mg/ml) 0 2 4 6 12 Control 0 0 5.01 5.12a 6.09a 7.03a 8.54a PIgG 0 4 5.00 4.93bc 5.51bc 6.26bc 8.59a Lactoferrin 10 Lactoferrin 0 4.99 5.13a 5.74b 6.41b 7.95b Lactoferrin-PIgG 10 Lactoferrin 4 5.01 5.04abc 5.56bc 6.03de 7.93b EDDA 0.1 EDDA . 0 5.04 5.04abc 5.41c 5.87e 5.02c EDDA-PIgG 0.1 EDDA 4 5.01 4.93c 5.07d 5.18f 4.84c HBED 0.1 HBED 0 5.01 5.08ab 6.06a 7.10a 8.80a HBED-PIgG 0.1 HBED 4 5.01 5.02abc 5.50bc 6.17cd 8.53a SE 0.017 0.028 0.051 0.039 0.054 Means in the same column bearing the same letter are not significantly different (P < 0.05). 89 Figure 14. The effect of lactoferrin, EDDA and HBED, with and without porcine immunoglobulins, on the Mean Log(CFU) mL-1 of E^ . c o l i 0 157 K88 (Experiment 2). THE EFFECT OF LACTOFERRIN ON THE GROWTH OF E. coli O 157 K88 | 3 | THE EFFECT OF EDDA ON THE GROWTH OF E. coli 0157 K88 90 EXPERIMENT 3 Introduction In Experiment 2 the antibacterial effects of lactoferrin, EDDA and HBED were tested in an in vitro system. Lactoferrin inhibited the growth of E^ c o l i 0 157 K88 over a 12 hour period. The large quantities of lactoferrin required to bind a l l the iron present in commercial milk replacers makes i t unsuitable as a feed additive. The synthetic iron chelators EDDA and HBED have low molecular weights and are inexpensive. This makes them candidates to replace lactoferrin in milk replacers. In Experiment 2, EDDA inhibited the growth of E^ . c o l i 0 157 K88 over a 12 hour period while HBED had no inhibitory effect. The objective of the present experiment was to test the effects of adding EDDA or HBED to milk re-placers fed to colostrum deprived piglets. Materials and Methods  Experimental Design The treatments used in Experiment 3 are shown in Table 18. The experiment was a 2 x 3 factorial layout with 2 levels of immunoglobulins and 3 levels of chelators. The piglets were reared in three outcome groups of 15,15 and 14 piglets. The outcome groups were started 28 days apart and an " a l l i n - a l l out" regimen was practiced. A l l piglets received the same diet on day 1. This was done to ensure that a l l piglets had approximately the same level of systemic PIgG. Any treatment differences would thus be due to the effect of PIgG and iron chelators in the small intestine. 91 Table 18. Experimental protocol for testing the effect of EDDA and HBED on the performance of a r t i f i c i a l l y reared piglets (Experiment 3 A). PIgG Chelator Treatment dav 1 dav 2-14 dav 2--14 POO8 20 0 0 PPO 20 4 0 POE 20 0 0.1 EDDA PPE 20 4 0.1 EDDA POH 20 0 0.1 HBED PPH 20 4 0.1 HBED A A l l values are in mg mL-1. 8 The f i r s t letter of the treatment name indicates the type of immunoglobulin received on day 1, the second letter indicates the type of immunoglobulin received on day 2-14 (P = porcine, 0 = none) the third letter indicates the type of chelator in the diet on days 2-14 (E = EDDA, H = HBED, 0 = none). The results were analyzed using the General Linear Models procedure of SAS (Statistical Analysis System Institute Inc. 1985) using the following least squares model. Yijk = U + Ci + Pj + Gn + CiPj + CiGk +PjGk + CiPjGk + Eijk Where Yijk = the dependent variable u = the overall mean 92 Ci = the effect of the i t h chelator P j = the effect of the j t h level of PIgG Gk = the effect of the k t h outcome group C JPj = the interaction between the i t h chelator and the j t h level of PIgG CiGk = the interaction between the i t h chelator and the k t h outcome group PjGk = the interaction between the j t h level of PIgG and the k t h outcome group CiPjGk = the interaction between the i t h chelator, the j t h level of PIgG and the k t h outcome group Eijk = the residual error for each sample Non-significant interactions were added to the error term and the results recalculated. Differences between means were analyzed by orthogonal contrasts. Animal Management Piglets were removed from the sow immediately after birth and fed as in Exp-eriment 1 with the one difference that the piglets were nipple fed by hand for 48 hours instead of 72 hours. The piglets adjusted easily to the automated feeder after 48 hours so the extra night of hand feeding was eliminated. Piglets were fed hourly for the f i r s t 6 hours then every 3 hours until 48 hours. From that point on the piglets were treated exactly as in Experiment 1. Piglets were injected with 100 mg of iron dextran on day 4 before blood samples were taken. Castration was performed on day 10. Iron Chelators EDDA and HBED were added as concentrated solutions to the liquid milk replacer before feeding. The f i n a l concentration of both iron chelators in the 93 milk replacer was 0.1 mg mL-1. The iron content of the dry milk replacer was 6.1 pg g - 1 on an as fed basis. The iron content of the water used to mix the milk replacer was 0.5 pg mL-1. This gave a iron content of 1.6 ug mL-1 of the liquid milk replacer as fed. The EDDA in the piglet diets was 10.3% saturated with iron. The HBED was 11.1% saturated with iron. Blood Samples Blood samples were taken from the orbital sinus with a IVt inch, 20 gauge needle. The blood was collected into heparinized 4 mL vacutainer tubes. One mL of blood was collected at birth and on day 1. Three mL of blood were collected at a l l other sampling times. Two microhematocrit tubes were f i l l e d with blood to determine packed c e l l volume. Another aliquot of blood was used to determine hemoglobin. The remainder was centrifuged at 500 x g for 15 minutes and the plasma was collected. The plasma was frozen until the end of the experiment when a l l other analyses were performed. Packed Cell Volume Microhematocrits were done on each blood sample on the day of collection. Canlab heparinized microhematocrit tubes were f i l l e d with blood and sealed with Critoseal. The tubes were then centrifuged for 15 minutes in a Canlab Interna-tional Microcapillary Centrifuge (Model MB) and read on a Canlab reader (Model CR). Hemoglobin Blood hemoglobin was measured by the cyanomethemoglobin method (Shoen and Solomon 1962). Duplicate analyses were done on the day of collection. Plasma Iron and Total Iron Binding Capacity Plasma iron and total iron binding capacity (TIBC) were measured on a Tech-nicon autoanalyzer (Model AA II). Plasma iron was measured using Technicon 94 method No. SE4-0025FL4. The procedure is based on that of Giovaniello et a l . (1967) and Stookey (1970). TIBC was measured by saturating the plasma trans-ferrin with iron and removing the excess iron with solid magnesium carbonate (Fielding 1980). The plasma was centrifuged and the supernatant was assayed for i t s iron content in the autoanalyzer. Results  Survival One piglet on the POO treatment died (see Table 19). The apparent cause of death was c o l i septicemia. A l l other piglets that died received diets containing HBED. This included 3 of 8 piglets on the POH diet and 2 of 7 piglets on the PPH diet. One piglet on the POH treatment died of E_j. c o l i septicemia. This piglet had an extremely high concentration of iron in i t s liver (1073 ug g-i compared to 175-472 ug g - 1 for other piglets from this experiment). Iron toxicity was also given as a possible cause of death. The cause of death of the other two piglets was reported as chronic dermatitis with marked hyperkeratosis, parakeratosis and acanthosis. There were no visible lesions of the liv e r , brain or kidney of these piglets and no specific evidence of toxicity. The apparent cause of death of the two piglets on the PPH diet that died was E. c o l i septicemia. One piglet also had non-hemolytic Staphylococcus aureus isolated from i t s spleen. Piglets receiving HBED had a significantly lower survival rate than those receiving either EDDA or no chelator. The survival rate for the POH treatment was significantly lower than the PPO, POE, PPE and PPH treatments. Diarrhea Significant differences in the level of diarrhea occurred only in the f i r s t week (see Table 20). Piglets receiving PIgG during days 2-14 treatments had less 95 diarrhea than those that did not. POE treatment piglets had a higher level of diarrhea than the combination of PPO, PPE and PPH. Average Daily Gains and Piglet Weights There were no significant treatment differences in the birth weights of the piglets (see Table 22). The Chelator x PIgG interaction was significant for average daily gains during weeks 2,3 and 4. Because of this only comparisons between treatment means were analyzed (see Table 21). The interaction was caused by the high average daily gains of the POE treatment compared to the POO and POH treatments. Piglets that received only EDDA during days 2-14 gained as well as piglets on the PPO, PPE and PPH treatments. The results were similar for mean piglet weights. The Chelator x PIgG inter-action was significant on days 21 and 28. The interaction was caused by the high weights of the POE treatment piglets relative to the POO and POH treatments. The only significant treatment differences occured on day 28. The piglets on the POH treatment weighed significantly less than the piglets on the POE, PPO, PPE and PPH treatments. Piglets that received only EDDA during days 2-14 weighed as much as piglets on the PPO, PPE and PPH treatments. Plasma Iron and Total Iron Binding Capacity The var i a b i l i t y for TIBC was high and no significant differences were found for any sampling time (see Table 23). Blood samples on day 4 were taken 2 to 4 hours after iron injections were given to the piglets. There was a rapid increase in TIBC following this injection. TIBC values ranged from 269-302 ug dLr 1 on day 1 and 574-666 pg d l r 1 on day 4. Plasma iron was not measured at birth or on day 1 (see Table 23). There were significant differences in plasma iron on day 14. Piglets that received no PIgG on day 2-14 had higher plasma iron concentrations than those that did receive 96 PIgG. The contrast POE vs PPO, PPE and PPH was also significant. The POE treatment had a higher plasma iron level than the other 3 treatments. Day 4 levels of plasma iron were markedly higher than the levels on day 7 due to the iron dextran injection. Packed Cell Volume and Hemoglobin Table 24 shows the packed c e l l volume and hemoglobin concentrations. On day 14 the packed c e l l volume of the piglets receiving no chelator was significantly higher than the combination of the EDDA and HBED treatment groups. There were no significant contrasts for hemoglobin on any sampling day. Plasma PIgG Table 25 and Figure 15 show the plasma PIgG levels. There were no significant differences between any of the means for which comparisons were made. Discussion The piglets in this experiment received 37.5 mg Kg body weight-1 day-1 of EDDA with no apparent i l l effects. Lambs receiving 50 mg Kg body weight - 1 day 1 showed definite signs of toxicity and 2 of 6 died (Stifel and Vetter 1967) . EDDA is not well absorbed when administered orally to rats (Hershko et a l . 1984b). This may also be the case for piglets. Poor absorption from the dige-stive tract would minimize systemic toxic effects. HBED is less toxic than EDDA. The LDso of HBED for rats was 800 mg Kg body weight-1 (Grady and Jacobs 1981). HBED was also poorly absorbed from oral doses given to rats (Hershko et a l . 1986b). The 37.5 mg Kg body weight-1 day-1 fed in the present study should have caused no problems. The survival of piglets receiving HBED was surprisingly low however. The severe dermatitis that affected the 2 piglets receiving the POH diet may have been caused by the HBED. The other piglet that died on this treatment 97 had an unusually high level of iron in i t s liver but there were no overt signs of toxicity except the skin lesions. The piglets on the PPH diet also had a lowered survival rate. They did not display any skin lesions but did show a greater susceptibility to E. c o l i . HBED, while not directly toxic, may interfere with the metabolism of iron and other metal ions in the piglet. Survival of the POO, PPO, POE and PPE treatment piglets was similar to the rates experienced in Experiment 1 for the porcine immunoglobulin treatment. The survival of piglets on POO and POE diets was not adversely affected by the absence of PIgG after day 1. The " a l l i n - a l l out" regimen used in this ex-periment allowed for a thorough cleaning of the experimental room among outcome groups of piglets. This prevented any build up of disease and meant that the level of environmental contamination was low as each outcome group began the experiment. The lack of a negative control group may also have contributed to the cleanliness of the environment. If a negative control group had been included in this experiment there may have been a higher level of disease and mortality. The diarrhea seen in this experiment was different from Experiment 1. The highest score observed was a 3 for one of the POH piglets. A l l other piglets had a maximum score of 2. This low level diarrhea was widespread however, and most piglets had a score of 2 at some point during the experiment. The " a l l i n -a l l out" regimen and the lack of a control group probably contributed to the low level of diarrhea seen in the present experiment. EDDA and HBED had no effect on diarrhea during the present experiment. In Experiment 2 i t was found that EDDA inhibited the growth of E;_ c o l i 0 157 K88 but this did not translate into lower incidence of diarrhea. Not a l l d i -arrhea i s caused by bacteria however. EDDA would probably have l i t t l e effect 98 on diarrhea caused by viruses. PIgG had a significant effect on the diarrhea scores during the f i r s t week of the experiment. PIgG would be effective against viruses and this may explain why i t was able to control diarrhea when EDDA could not. During the second week, diarrhea scores were the same whether the piglets received PIgG on days 2-14 or not. During the f i r s t week a l l piglets received the same diet on day 1. The effect of this was to minimize treatment differences in average daily weight gain during the f i r s t week. During week 2, the beneficial effect of feeding immuno-globulins in the diet after day 1 was manifest in the increased average daily weight gains of the PPO, PPE and PPH treatments compared to the POO and the POH diets. Piglets on the POE diet also had higher average daily weight gains than the POO and POH treatment piglets. POH piglets had a mean weight of only 5,090 g on day 28. POO piglets also had a relatively low mean weight of 5,750 g. POE piglets weighed 6,240 g on average, as high as for any other treatment. There was no synergistic effect when EDDA and PIgG were fed together. The PPE treat-ment had weight gains and piglet weights almost identical to the POE treatment except during week 1. During that period the POE treatment piglets had lower average daily gains than those on the PPE treatment. These findings are similar to those in Experiment 2 but are not, however, directly comparable. In Experiment 2, bacterial inhibition was measured in a closed system. The numbers of bacteria, the nutrients available, immunoglobulins and iron chelators were fixed. In Experiment 3, these components were steadily entering and leaving the system, the system being the gastrointestinal tract of the piglet. EDDA appears to have an effect as beneficial as that of PIgG in the small intestine of the piglet. The average daily gains of the POH treatment piglets were worse than for POO treatment piglets. During weeks 1 and 2 the average daily gains for POH were comparable to the other treatments. During weeks 3 and 4 this treatment produced a marked depression in rate of weight gain. PPH piglets did not show this depression in weight gains during weeks 3 and 4. The absence of PIgG coupled with a possible toxicity from HBED may have been a greater challenge than the POH piglets could withstand. This could have led to the lower survival and weight gains. The PPO treatment and the porcine immunoglobulin treatment from Experiment 1 are identical treatments. Average daily gains of the piglets in the present experiment were lower in the f i r s t week than in experiment 1 (90.8 vs 153.4 g day-1). This difference disappeared in subsequent weeks and the 28 day piglet weights were slightly higher in for Experiment 3 than Experiment 1. The total iron binding capacity (TIBC) for the piglets measured agrees with other studies (Furugouri, 1971;1972;1973). The TIBC was low at birth and remained low until after the iron dextran injection. After the iron dextran injection, the TIBC rose about 400 pg dLr 1 in the few hours. None of the con-trasts were significant for TIBC on any day. EDDA and HBED did have significant effects on plasma iron. EDDA and HBED piglets had higher plasma iron levels on day 14 than those that received no chelator. This may have been due to the EDDA and HBED in the blood. Any c i r -culating EDDA or HBED would almost certainly be bound to iron and elevate plasma iron. A small quantity of the circulating chelators could affect plasma iron. The elevated plasma iron would not affect TIBC. Furugouri (1971) found that TIBC was independent of the plasma iron concentration. PIgG also affected plasma iron on day 14. Piglets receiving PIgG on days 100 2-14 had lower plasma iron concentrations than those that received no PIgG on days 2-14. This is opposite of what would be expected. The PIgG in the diet con-tained some iron so that piglets receiving PIgG on days 2-14 would obtain more dietary iron than those that received no PIgG. One might also expect an equal or higher infection rate in piglets not receiving PIgG on days 2-14. Infection causes plasma iron to decrease (Van Snick et a l . 1974). Piglets not receiving PIgG should therefore have an equal or lower plasma iron concentration. Plasma iron was not measured during the f i r s t two days after birth. After the injection of iron dextran on day 4, plasma iron was about 400 ug d l r 1 for a l l treatments. The level decreased to less than 200 on day 7. The concentration on day 21 was about the same as the plasma iron levels found in other studies Furugouri (1971). The contrast No chelator vs EDDA and HBED" was significant on day 14 for packed c e l l volume. The difference disappeared by day 21. EDDA and HBED may interfere with the synthesis of red blood cells but the effect was small and transitory. The packed c e l l volumes of piglets in this experiment were similar to those found in other studies for the f i r s t 2 weeks (Zimmerman et a l . 1959, Miller et a l . 1961, Kay et a l . 1980). For weeks 3 and 4 the values were 5-7% lower than those reported for naturally reared piglets. None of the contrasts were significant for hemoglobin concentration on any day. The values found in this experiment were similar to those found by Miller et a l . (1961) but 2-3 g d l r 1 higher than those reported by Kay et a l . (1980) throughout the experimental period. Plasma PIgG levels were similar for a l l treatments. This was not unusual con-sidering a l l piglets received the same diet on day 1. The difference in plasma PIgG ranged from 19.6 to 25.5 mg mL-1 on day 1. This range was much smaller on 101 day 4 (18.2-21.8). Most treatment means were at a minimum on day 21. The d i -fference between the means on day 21 and 28 was not significant. The values were similar to those in Experiment 1 for the Porcine treatment group. In conclusion, HBED has no potential as a lactoferrin substitute in milk replacers for colostrum deprived piglets. It had no antibacterial effect in vitro and decreased survival and growth rates when fed to piglets. EDDA on the other hand appears to have promise as an additive to sow milk replacers. It has significant antibacterial properties in vitro and did not have any negative effects on piglet performance when included in the diet. 102 Table 19. The effect of EDDA or HBED, with or without porcine immunoglobulins on the survival of colostrum deprived piglets (Experiment 3). Number Surviving at day Birth 7 14 21 28 Treatment N N N N N % SE POO 8 8 8 7 7a 88 12.0 PPO 7 7 7 7 7a 100 12.9 POE 7 7 7 7 7d 100 12.9 PPE 7 7 7 7 7d 100 12.9 POH 8 7 7 6 5b 62 12.0 PPH 7 7 6 6 5 70 13.1 N = the number of surviving piglets. a The contrast POO and PPO vs POE, PPE, POH and PPH is significant (P < 0.05) b The contrast POH vs PPO, POE, PPE and PPH is significant (P < 0.05) d The contrast POE and PPE vs POH and PPH is significant (P < 0.05) 103 Table 20. The effect of EDDA or HBED, with or without porcine inrrmmoglobulins on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 3). week Treatment 1 2 3 4_ POO 1.6e 1.2 1.3 1.1 PPO 1.2 1.1 1.1 1.1 POE 1.7ce 1.3 1.4 1.0 PPE 1.3 1.2 1.3 1.1 POH 1.5e 1.2 1.2 1.0 PPH LJ2 h2 1.1 1.0 c The contrast POE vs PPO, PPE and PPH is significant (P < 0.05) e The contrast POO, POE and POH vs PPO, PPE and PPH is significant (P < 0.05) 104 Table 21. The effect of EDDA or HBED, with or without porcine innunoglobulins on the average daily gains of colostrum deprived piglets (Experiment 3). Average Daily Gains (g day^ -) Treatment week 1 week 2 week 3 week 4 X SE X SE ' X SE X SE POO 80 10 150 12 220 21 190 20 PPO 90 10 170 12 230 21 210 20 POE 80 10 160 12 240 21 230 20 PPE 110 10 150 12 240 21 220 20 POH 90 12 150 14 180 25 120b 23 PPH 100 12 160 14 220 25 200 24 b The contrast POH vs PPO, POE, PPE and PPH is significant (P < 0.05) 105 Table 22. The effect of EDDA or HBED, with or without porcine inrauncglobulins on the body weights of colostrum deprived piglets (Experiment 3). Body Weights (q) Treatment Birth dav • 7 dav 14 dav 21 dav 28 X SE X SE X SE X SE X SE POO 1,310 69 1,830 70 2,860 119 4,430 207 5,750 289 PPO 1,250 74 1,900 70 3,110 119 4,750 207 6,230 287 POE 1,220 74 1,820 70 2,950 119 4,630 207 6,240 289 PPE 1,360 74 2,020 71 3,050 121 4,720 211 6,240 294 POH 1,270 69 1,880 82 2,950 140 4,250 244 5,090 341 PPH 1,290 74 1,990 84 3,110 143 4,660 249 6,060 347 Contrasts - - - - b b The contrast POH vs PPO, POE, PPE and PPH is significant (P < 0.05) 106 Table 23. The effect of EDM or HBED, with or without porcine immunoglobulins on the plasma iron and total iron binding capacity (TIBC) of colostrum deprived piglets (Experiment 3). Age of Piglets (days) 1 3 1 4 7 14 21 28 Treatment X SE X SE X SE X SE X SE X SE X SE Plasma Iron (ug dlr1) POO na na 440 24 173 15 138ae 12 97 9 69 3 PPO na na 439 26 182 17 114a 13 95 9 69 4 POE na na 428 29 156 19 174ce 15 93 11 65 4 PPE na na 403 22 192 14 132 11 101 8 73 3 POH na na 441 26 188 17 153e 13 87 9 70 4 PPH na na 392 26 181 17 134 13 92 9 73 4 TIBC (ug dlr1) POO 249 23 294 22 666 32 526 61 406 53 548 22 588 24 PPO 269 25 281 24 654 35 541 68 411 59 535 24 570 26 POE 291 21 302 20 643 29 484 57 442 50 544 21 570 22 PPE 268 23 273 22 598 32 455 61 434 53 562 22 543 24 POH 231 26 269 25 574 37 423 71 494 62 548 26 555 27 PPH 263 26 291 25 587 37 504 71 442 62 569 26 654 27 a The contrast POO and PPO vs POE, PPE, POH and PPH is significant (P < 0.05) c The contrast POE vs PPO, PPE and PPH is significant (P < 0.05) e The contrast POO, POE and POH vs PPO, PPE and PPH is significant (P < 0.05) 107 Table 24. The effect of EDDA or HBED, with or without porcine inraunoglobulins on the packed cell volumes and hemoglobin concentrations of colostrum deprived piglets (Experiment 3). Age of Piglets (days) 0 1 4 7 14 21 28 Treatment X SE X SE X SE X S E X S E X S E X S E Packed Cell Volume (%) POO 34.3 2.5 29.2 3.7 30.2 3.1 31.0 2.0 32.0a 1.7 32.5 2.7 30.0 2.5 PPO 32.5 2.5 24.2 3.7 27.7 3.1 28.8 2.0 31.3a 1.7 31.1 2.7 28.8 2.5 POE 33.4 1.9 26.5 2.9 25.7 2.4 27.7 1.6 28.0 1.3 30.4 2.1 29.7 1.9 PPE 31.2 2.1 23.9 3.2 24.4 2.7 28.0 1.7 29.2 1.5 33.0 2.4 31.4 2.2 POH 29.4 2.5 26.1 3.7 24.5 3.1 28.8 2.0 29.5 1.7 29.3 2.7 29.4 2.2 PPH 35.3 2.5 26.5 3.7 28.8 3.1 27.0 2.0 29.3 1.7 31.8 2.7 28.6 2.5 BBCGLCBIH (g d l / ' ) POO 13.6 1.4 12.0 1.1 11.0 1.1 11.0 0.8 11.6 0.8 11.7 1.1 12.1 1.1 PPO 11.8 1.6 10.4 1.3 10.7 1.3 10.6 1.0 12.6 0.9 11.3 1.3 11.0 1.3 POE 12.1 1.3 10.5 1.0 10.2 1.0 9.3 0.7 10.6 0.7 11.4 1.0 10.7 1.0 PPE 13.2 1.3 10.3 1.0 8.8 1.0 9.2 0.7 11.9 0.7 12.1 1.0 11.0 1.0 POH 11.7 1.7 10.0 1.3 8.8 1.3 9.9 1.0 11.9 0.9 11.1 1.3 11.6 1.3 PPH 11.8 1.7 10.3 1.3 11.4 1.3 10.0 1.0 11.3 0.8 11.4 1.1 9.5 1.1 a The contrast POO and PPO vs POE, PPE, POH and PPH is significant (P < 0.05) 108 Table 25. The effect of EDDA or HBED, with or without porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 3)1. Age of Piglets (days) 0 1 4 7 14 21 28 Treatment X SE X SE X SE X SE X SE X SE X SE POO <1.3 - 19.6 3.0 21.6 2.4 17.0 2.4 9.5 1.6 9.0 1.1 7.8 1.5 PPO <1.3 - 20.3 3.0 21.4 2.4 14.0 2.4 8.6 1.6 6.9 1.1 8.2 1.5 POE <1.3 - 25.5 3.0 21.9 2.4 15.8 2.4 7.6 1.6 7.2 1.1 8.5 1.5 PPE <1.3 - 19.6 3.0 18.2 2.4 13.6 2.4 8.3 1.6 7.4 1.1 7.2 1.5 POH <1.3 - 20.8 3.0 18.5 2.4 15.9 2.4 10.6 1.6 8.6 1.1 8.7 1.5 PPH <1.3 - 22.3 3.3 21.8 2.7 15.8 2.6 10.1 1.7 7.7 1.2 7.3 1.6 1 There were no significant contrasts (P < 0.05). Figure 15. The effect of EDDA or HBED, with or without porcine immunoglobulins on the plasma immunoglobulin concentrations of colostrum-deprived piglets (Experiment 3). For s t a t i s t i c a l analysis see Table 25. 110 EXPERIMENT 4 Introduction Providing passive immunity to a r t i f i c i a l l y reared piglets can be divided into two stages. The colostral stage occurs during approximately the f i r s t 24 hours of l i f e . During this time immunoglobulins and other protective factors are absorbed directly into the blood stream of the piglet. This provides systemic immunity and regulates the immune system of the piglet. The milk stage starts after the colostral stage ends and lasts until the piglet i s weaned from the protective factors in milk. These factors prevent enteric infections by pre-venting bacterial growth and adhesion within the intestinal tract. Experiment 1 demonstrated that during the colostral stage, porcine immuno-globulins are required. Plasma IgG concentrations averaged 20.8 mg mL-1 on day 1 when PIgG was fed. When BIgG was fed plasma IgG levels averaged 4.0 mg mL-1 on day 1. During the f i r s t week of l i f e , piglets receiving PIgG gained weight at a greater rate than those that received BIgG. There was also a trend to higher survival among piglets receiving PIgG instead of BIgG. During the milk stage, the level of diarrhea in piglets receiving BIgG was not significantly different from those fed PIgG. This may indicate that BIgG can replace PIgG during the milk stage. In Experiment 3, the addition of EDDA to milk replacers during the milk stage had a beneficial effect on piglet growth. The objective of Experiment 4 was to compare the effects of adding BIgG, PIgG and/or EDDA to milk replacers to see which factor or combination of factors provided the best passive protection to the piglet during the milk stage. Materials and Methods The treatments used in Experiment 4 are shown in Table 26. 111 Table 26. Experimental protocol for studying the administration of EDDA with porcine or bovine immunoglobulins (Experiment 4). Treatment dav 1 dav 2-•14 dav 15--28 la EDDA Icr EDDA Ig EDDA Control 0 0 0 0 0 0 BBO 25 BIgG 0 5 BIgG 0 0 0 POO 25 PIgG 0 0 0 0 0 POE 25 PIgG 0.1 0 0.1 0 0 PBO 25 PIgG 0 5 BIgG 0 0 0 PBE 25 PIgG 0.1 5 BIgG 0.1 0 0 PPO 25 PIgG 0 5 PIgG 0 0 0 PPE 25 PlaG 0.1 5 PIgG 0.1 0 0 A l l values are in mg mL-1 The concentration of immunoglobulins in the diets was increased from 20 mg mL- 1 on day 1 and 4 mg mL-1 on days 2-14 to 25 mg mL-1 and 5 mg mL-1. These are closer to the 26.7 and 13.3 mg mL-1 levels recommended by McCallum (1977). The levels were increased for 2 reasons. F i r s t l y , to see i f any increase in growth rate occurred at these higher levels. Secondly, to see what effect the increased dietary immunoglobulins on day 1 would have on the level of plasma immunoglob-ulins. In Experiment 3, plasma immunoglobulin levels actually increased from day 1 to day 4 for some treatments. Leary and Lecce (1979) proposed that immunoglobulin specific receptors on enterocytes are responsible for the selective absorption of PIgG from the diet into the bloodstream. It was hypothesized that 20 mg mL-1 1 2 1 of PIgG in the diet was inadequate to saturate these receptors. This meant that the absorption of PIgG continued after 24 hours thus giving higher plasma PIgG values on day 4. By increasing the level of PIgG to 25 mg mL-1 i t was hoped that the specific immunoglobulin receptors would be saturated within the f i r s t 24 hours of l i f e and gut closure would occur earlier. The piglets were raised in 5 outcome groups. The outcome groups were farrowed less than 4 weeks apart so that newborn piglets entered the experimental room with the piglets from the previous outcome group s t i l l present in the room. The results were analyzed using the General Linear models procedure of SAS using the following least squares model. Yu = u + Ti + Gj + TiGj + Eij Where Ytj = the dependent variable u = the overall mean Ti = the effect of the i t h treatment Gj = the effect of the j t h outcome group TiGj = the interaction between the i t h treatment and j t h outcome group Eij = the residual error for each sample Non-significant interactions were added to the error term and the results recalculated. Differences between the means were analyzed using orthogonal con-trasts. A l l other experimental methods and procedures were as described for Ex-periment 3. Results  Survival The survival of the Controls was 0 out of 9 (see Table 27}. This was 113 significantly lower than for any other treatment. A l l piglets were dead before day 7. The cause of death for 8 of the 9 was septicemia. The major organism responsible was E^ . c o l i which was isolated from the tissues of every Control piglet that died, a hemolytic Staphylococcus spp. , fi hemolytic Streptococcus  spp., Klebsiella spp. and Actinobacillus suis were also isolated from some Control piglets. Staphylococcus spp. and Streptococcus spp. were found in Controls in every outcome group. Actinobacillus suis was found in the fourth outcome group and Klebsiella spp. was isolated from both Controls from the f i f t h outcome group. The remaining piglet died of bronchopneumonia caused by E. c o l i . Four of 10 BBO treatment piglets died. This was significantly more than for any other treatment except the Controls. Bronchial pneumonia caused by E^ . c o l i was the cause of death for the 2 piglets that died before 14 days of age. The two piglets that died during week 3 died of Streptococcus suis II infection. There were no significant differences in survival rates between any of the treatments that received PIgG on day 1. Streptococcus suis II infection was re-sponsible for the death of the piglets on the PBO and the PBE treatments. One Porcine treatment piglet died. The cause was omphalitis (Kidney infection) caused by E^ . c o l i . Diarrhea The Control treatment had an average weekly diarrhea score of 3.4 during the f i r s t week (see Table 28). This was significantly higher than for any other treatment. There were no diarrhea scores for weeks 2-4 since a l l Controls were dead after the f i r s t week. The type of immunoglobulins fed during week 1 had a significant effect on the amount of diarrhea seen. Piglets that received no immunoglobulins during 114 week 1 had more diarrhea than those that received BIgG or PIgG. Piglets re-ceiving BIgG in week 1 had significantly more diarrhea than those receiving PIgG. The PPO treatment piglets had significantly less diarrhea than those on the BBO treatment. EDDA had no significant effect on diarrhea during week 1. During week 2, piglets receiving no immunoglobulins had more diarrhea than piglets receiving PIgG or BIgG. No other effects were significant during weeks 2 to 4. There was an increase in diarrhea from week 2 to week 3 for groups re-ceiving BIgG or PIgG during weeks 1 and 2. Average Daily Gains and Piglet Weights Tables 29 and 30 show average daily gains and piglet weights respectively. The piglet weights for days 7, 14, 21 and 28 and a l l average daily gains for the Control piglets were non-estimable. None of the comparisons between treatments for average daily gain were significant during week 1. During week 2 the piglets receiving EDDA had lower average daily gains than those not receiving EDDA. Also during week 2, piglets that received no immunoglobulins had lower average daily gains than those that received either BIgG or PIgG. The BBO treatment piglets had lower average daily gains during week 3 than the piglets that received PIgG on day 1. On days 21 and 28, mean piglet weights for the BBO treatment were lower than for piglets that received PIgG on day 1. Piglets that received EDDA in the diet had lower mean weights than those not receiving EDDA on days 14, 21 and 28. Also on days 14, 21 and 28 piglets that received no immunoglobulins on days 2-14 had lower weights than piglets that received either BIgG or PIgG on days 2-14. Plasma Iron and Total Iron Binding Capacity Table 31 shows the plasma iron and TIBC of the piglets. The Controls had 115 significantly lower plasma iron and TIBC on day 4 than for a l l other treatments. The BBO treatment piglets had significantly lower plasma iron levels on days 7 and 14 than a l l other treatments. The treatments receiving EDDA had significantly higher plasma iron levels on day 7. On days 14 and 21, plasma iron was higher in piglets that received no immunoglobulins on days 2-14 than in piglets that received either BIgG or PIgG. The same was true for TIBC on day 1. Packed Cell Volume and Hemoglobin The Controls had a higher packed c e l l volume on day 4 than any other treat-ment (see Table 32). The only other significant effect on packed c e l l volume and hemoglobin was due to EDDA. Piglets that were fed EDDA had significantly lower packed c e l l volumes on days 1, 7 and 14. They also had significantly lower hemoglobin levels on days 1 and 14 (see Table 32). Plasma Immunoglobulins Table 33 and Figure 16 show the plasma immunoglobulin levels. PIgG and BIgG were measured for the BBO treatment piglets. PIgG only was measured for a l l other treatments. There were no significant differences in PIgG levels for any of the treatments receiving PIgG during the f i r s t 24 hours of l i f e . None of the Control piglets had measurable levels of PIgG during the course of the experiment. The total of PIgG and BIgG concentrations in the plasma of the BBO treatment piglets was compared to the PIgG levels in the other treatments. The BBO piglets had significantly lower total IgG than the other treatments on days 1,4,7 and 14. Plasma PIgG was measurable on day 7 and rose to 10.1 mg mL-1 by day 21. Discussion The survival of the Control group was 0%. Furthermore a l l piglets died 116 before 7 days of age. Common environmental pathogens such as c o l i . Strepto- coccus spp. and Staphylococcus spp. were present throughout the experiment. Other less common pathogens such as Klebsiella spp. and Actinobacillus suis became apparent in later outcome groups. Most Controls died of septicemia caused by one or more of the above pathogens. The pattern here is similar to that found in Experiment 1. Experiment 3 had no Control treatment so that no comparisons are possible. The BBO treatment piglets had a survival rate of 60%. This is slightly lower than for the bovine immunoglobulin fed group in Experiment 1 even though the levels of BIgG were higher in the present experiment. The lower survival rate of both the Control and BBO piglets may indicate a higher level of contagion than in Experiment 1. The room was in continuous use in this experiment for over 4 months compared to 2 months for Experiment 1. In Experiment 3 an " a l l i n - a l l out" regimen was practiced. The level of environmental pathogens may have been lower but with no Control group comparisons are d i f f i c u l t . The BBO piglets died of different causes than the Controls. E_j_ c o l i Bronch-opneumonia was the cause of death of two piglets which died on days 7 and 14. Streptococcus suis II septicemia was the cause of death of the other two BBO piglets. They died on days 16 and 20. The BBO piglets appeared able to fend off E^ c o l i septicemia successfully but they were susceptible to subsequent infections by other organisms. The piglets that received PIgG on day 1 had a very low mortality rate. Only 3 of 60 piglets died. Streptococcus suis II was responsible for the deaths of the one PBO and one PBE piglet that died during week 3. The one PPO piglet that died on day 2 of L c o l i omphalitis weighed only 800 g at birth and had a day 1 plasma PIgG level of only 8.5 mg mL-1. Mortality rates in Experiment 3 were 117 similar to those in the present experiment for analogous treatments. Both experiments had POO, PPO, POE and PPE treatments. The mortality rate for those four treatments was 1 out of 29 in Experiment 3 and 1 out of 40 in the present experiment. The Controls in the present experiment had a week 1 diarrhea score of 3.4 compared to 3.1 for Controls in Experiment 1. The BBO piglets had week 1 diar-rhea scores of 1.5 compared to 1.0 for the analogous Bovine Ig treatment in Experiment 1. This also points to a higher level of environmental pathogens in the present experiment. The addition of EDDA to BIgG or PIgG did not produce a marked decrease in the level of diarrhea during week 1. In the absence of immunoglobulins however, EDDA was able to decrease the level of diarrhea during the f i r s t week of l i f e . During week 1, the POO treatment piglets had a diarrhea score of 1.7 compared to 1.4 for the POE piglets. This effect was not observed in Experiment 3. Porcine immunoglobulins for two weeks after birth produced lower levels of diarrhea than either BIgG or No immunoglobulins. This i s similar to the result found in Experiment 3 where after day 14, the diarrhea scores went up for the PBO, PBE, PPO and PPE treatments. The piglets on these treatments were weaned from dietary immunoglobulins on day 14. This may have caused the increase in diarrhea. The increase in was moderate however. Most piglets experienced a few days of a diarrhea score of 2 and then returned to normal. The average daily weight gains for the BBO piglets were low until week 4. The difference was only significant during week 3 but the effect of 3 weeks of low average daily gains gave the BBO treatment piglets the lowest day 28 weight of any treatment group. This is different from the results for Experiment 1 where the bovine immunoglobulin fed piglets had depressed average daily gains 118 only during week 1. The reason for this may be an increased level of environ-mental pathogens in this experiment. EDDA in this experiment resulted in decreased average daily gains, and piglet weights. This i s in direct contrast to the results of Experiment 3 where EDDA had a positive effect on piglet weights and average daily gains. The only difference between Experiment 3 and 4 is that piglets received EDDA on day 1 in Experiment 4. EDDA is not well absorbed from oral doses by rats (Hershko et a l . 1984b). This may also be the case for piglets after gut closure. Before gut closure takes place, EDDA may be absorbed directly into the blood stream along with immunoglobulins. Extra absorption of EDDA on the f i r s t day of l i f e increases i t s toxic effects. EDDA should not be fed until after gut closure takes place. In Experiment 1 and the present experiment i t was noted that on day 1, PIgG is not replaceable by BIgG. The treatments receiving no immunoglobulins (POO and POE) during days 2-14 had inferior average daily gains. The piglet weights for these two treatments were low on days 14,21 and 28. There is a significant benefit to continued feeding of immunoglobulins after day 1. The day 28 weights for the PBO and PPO treatments were not significantly different. Both of these treatments received PIgG on day 1 but the PBO piglets got BIgG on days 2-14 while the PPO piglets received PIgG. After day 1, BIgG and PIgG are equally effective in promoting piglet growth. This may be due to an overlap in the anti-gens expressed by bovine and porcine enteric pathogens. This would lead to an overlap in the immunoglobulin speci f i c i t i e s found in bovine and porcine sera and give the results seen here. Depending on the cost of producing PIgG and BIgG, either may be used to fo r t i f y piglet diets after day 1. The Controls had a plasma iron concentration about 100 pg dLr 1 lower than 119 the other treatments on day 4. Bacterial infection causes a decrease in plasma iron (Cartwright et a l . 1946). Since the Controls experienced more bacterial infections than any other group, i t i s not surprising that their plasma iron levels were lower. The piglets received 100 mg of iron dextran on day 4. Six of the 9 Controls died on days 4 or 5. Iron increases the ab i l i t y of bacteria to grow in host tissues (Weinberg 1984). The iron dextran may have hastened the death of the Controls by increasing the availability of iron to bacteria. Knight et a l . (1984) found that 100 mg of iron dextran did not compromise the piglet's a b i l i t y to limit the availability of iron to bacteria. The experiment involved healthy 2 week old piglets, however. The Control piglets in this experiment may not have been able to synthesize enough lactoferrin and transferrin to chelate the injected iron and sequester i t in the l i v e r . Evidence that this was the case can be seen in the day 4 TIBC of the Controls. It was significantly lower than for any other group suggesting an ina b i l i t y to rapidly synthesize iron-binding proteins. The plasma iron of the BBO treatment piglets was low on days 7 and 14. This treatment also suffered from a high level of bacterial infections. Presumably this was the cause of the low plasma iron concentrations in these piglets. The piglets receiving EDDA had high plasma iron levels on day 7. This is similar to the result in Experiment 3 where EDDA and HBED had higher plasma iron levels on day 14. This may be caused by EDDA bound iron present in the plasma. The piglets receiving No Ig had higher plasma iron levels than those re-ceiving BIgG or PIgG on days 14 and 21. The same result occurred in Experiment 3 on day 14. It is d i f f i c u l t to envision the mechanism that causes this effect. 120 Packed c e l l volumes and hematocrits for the Controls were high on day 4. This is possibly due to dehydration. Most Controls had severe diarrhea on day 4. This may have caused dehydration severe enough to elevate these blood measures. Piglets receiving EDDA had depressed packed c e l l volumes on days 1,7 and 14. They also had depressed hemoglobin values on days 1 and 14. This effect was present in Experiment 3 to a lesser extent. Packed c e l l volumes were low only on day 14 in that experiment and hemoglobin was not affected. The larger effect in the present experiment may be due to the increased absorbtion of EDDA on day 1. The effect on packed c e l l volume and hematocrit disappeared when EDDA was removed from the diet. In Experiment 4 the level of PIgG and BIgG included in the diets was increased to 25 mg mL-1 on day 1 and 5 mg mL-1 on days 2-14. This resulted i n an increase in plasma PIgG concentrations compared to those in the previous experiments. Day 1 plasma PIgG values ranged from 19.6 to 25.5 mg mL-1 i n Experiments 1 and 3. In the present experiment, values ranged from 24.6 to 29.5 mg mL-1. This increase in plasma PIgG is nearly proportional to the increase in dietary PIgG. The day 4 plasma PIgG levels were lower than the day 1 levels for a l l treatments. This suggests that gut closure occurred within the f i r s t 24 hours after birth. Experiment 1 concluded that the best level of PIgG on day 1 would provide adequate immune protection and minimize the inhibition of active immunity. The cr i t e r i a for "adequate" immunity are survival, diarrhea and weight gain. Com-paring analogous treatments from Experiments 1,3 and 4 (i.e. PIgG on day 1, PIgG on days 2-14) i t can be seen that survival did not change when the level of PIgG in the diet was increased. Diarrhea, however, may have decreased slight-ly in the present experiment. Diarrhea in the f i r s t two experiments ranged from 121 1.1 to 1.2 during the f i r s t two weeks. In the present experiment i t was 1.0 for the f i r s t two weeks. The level of environmental pathogens may have been higher in the present experiment than in Experiments 1 and 3. In Experiment 1, the room was in continuous use for a shorter period of time than in the present experiment. In Experiment 3, an " a l l i n - a l l out" regimen was used. In the present experiment, the room was in continuous use for a longer period and an " a l l i n - a l l out" regimen was not used. The Controls had 100% mortality. In spite of this, the weight gains of the piglets in this experiment were slightly higher than in Experiments 1 and 3. Day 28 piglet weights for Experiments 1 and 3 averaged 5,980 and 6,060 g respectively. The average day 28 weights in the present experiment were 6,350 g. The birthweights of pigs in the 3 experiments were virtually the same. The increased level of dietary PIgG resulted in an increase of about 300 g liveweight over 28 days. Based on a l l these c r i t e r i a , the level of PIgG fed in the current experiment is better than the lower level used in the previous experiments. The plasma PIgG remained higher than in the previous experiment throughout the 28 day period of the experiment. By day 28 the difference was negligible. The previous experiments ranged-from 7.2 to 9.5 mg mL-1 on day 28. In the present experiment values ranged from 8.7 to 10.1 mg mL-1. The BBO treatment piglets also received higher levels of BIgG on day 1 than the analogous treatment in Experiment 1. In Experiment 1 the day 1 plasma BIgG was 4.0 mg mL-1. In the present experiment the day 1 value was 5.7 mg mL-1. Increasing the level of BIgG in the diet resulted in a nearly proportional increase in plasma BIgG. The plasma BIgG was s t i l l poorly absorbed compared to PIgG. It also provided inadequate immune protection. The active synthesis of PIgG by the BBO treatment piglets was measurable by 122 day 7. This is similar to the result obtained in Experiment 1. The day 21 and 28 PIgG values for the BBO treatment piglets were slightly lower than the levels found in Experiment 1 but not markedly so. In conclusion, EDDA should not be fed until after gut closure takes place. EDDA had no deleterious effect on piglet growth when fed on days 2-14 but when fed on days 1-14 significant reductions in piglet growth rates occurred. There is some question whether EDDA should be fed to colostrum deprived piglets at a l l . Immunoglobulins alone supported piglet growth rates that were as good as EDDA alone or EDDA with porcine immunoglobulins. This coupled with EDDA's potential toxicity may make i t unsuitable as an additive to sow milk replacers. Bovine immunoglobulins are not well absorbed from the diet during the f i r s t day of l i f e and porcine immunoglobulins must be fed to ensure adequate passive systemic immunity. After day 1, however, either bovine or porcine immunoglobu-lins may be used to provide local immune protection in the intestinal tract. 123 Table 27. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the survival of colostrum deprived piglets (Experiment 4). Number Surviving at day Birth 7 14 21 281 Treatment N N N N N % SE Control 9 0 0 0 0a 0 9.8 BBO 10 9 8 6 6b 60 9.2 POO 10 10 10 10 10 100 9.3 POE 10 10 10 10 10 100 9.4 PBO 10 9 9 9 9 90 9.3 PBE 10 10 9 9 9 93 9.4 PPO 10 9 9 9 9 88 9.8 PPE 10 10 10 10 10 100 9.2 N = the number of surviving piglets a The contrast "Controls vs all other treatments" is significant (P < 0.05) b The contrast "BBO vs POO, POE, PBO, PBE, PPO and PPE" is significant (P < 0.05) 124 Table 28. The effect of EDDA, with and without bovine or porcine iinmunoglobulins, on the average weekly diarrhea scores of colostrum deprived piglets (Experiment 4). week Treatment 1 2 3 4 Control 3.4a Non-est1 Non-est Non-est BBO 1.5 1.2 1.4 1.2 POO 1.7d 1.4d 1.3 1.1 POE 1.4d 1.4d 1.4 1.1 PBO 1.3e 1.1 1.3 1.1 PBE 1.3ef 1.1 1.3 1.1 PPO 1.0 1.0 1.2 1.1 PPE 1.1 1.1 1.3 1.1 1 Non-est = non-estimable a The contrast "Controls vs all other treatments" is significant (P < 0.05) d The contrast "POO and POE vs PBO, PBE, PPO and PPE" is significant (P < 0.05) e The contrast 'TB0 and PBE vs PPO and PPE" is significant (P < 0.05) f The contrast "PBO vs PPO" is significant (P < 0.05) 125 Table 29. The effect of EDDA, with and without bovine or porcine iimnunoglobulins, on the average daily gains of colostrum deprived piglets (Experiment 4). Average Daily Gains (q day!) week 1 week 2 week 3 week 4 Treatment X SE X SE X SE X SE Control Non-est1 Non-est Non-est Non-est BBO 90 9 100 13 120b 19 200 16 POO 80 7 llOd 11 200 15 190 13 POE 100 7 90cd 11 200 15 210 12 PBO 100 8 160 11 240 16 230 13 PBE 80 8 110c 11 220 16 210 13 PPO 110 8 180 12 240 17 210 14 PPE 80 8 120c 11 190 —•• 16 220 13 1 Non-est = non-estimable b The contrast "BBO vs POO, POE, PBO, PBE, PPO and PPE" is significant (P < 0.05) c The contrast "POO, PBO and PPO vs POE, PBE and PPE" is significant (P < 0.05) d The contrast "POO and POE vs PBO, PBE, PPO and PPE" is significant (P < 0.05) 126 Table 30. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the body weights of colostrum deprived piglets (Experiment 4). Body Weights (g) Treatment Birth dav 7 dav 14 dav 21 day 28 X SE X SE X SE X SE X SE Control 1,160 71 Non-est1 Non-est Non-est Non-est BBO 1,180 67 1,810 65 2,520 130 3,370b 202 4,750b 267 POO 1,060 67 1,730 52 2,500d 104 3,910d 161 5,260d 213 POE 1,210 67 1,850 51 2,480cd 102 3,900cd 157 5,390cd 207 PBO 1,090 67 1,910 54 2,990 108 4,640 167 6,240 221 PBE 1,270 68 1,750 54 2,520c 109 4,040c 168 5,510c 222 PPO 1,260 71 1,970 58 3,230 116 4,890 180 6,350 237 PPE 1,160 67 1,770 53 2.610c 106 3,950c 165 5,470c 218 1 Non-est = non-estimable b The contrast "BBO vs POO, POE, PBO, PBE, PPO and PPE" is significant (P < 0.05) c The contrast "POO, PBO and PPO vs POE, PBE and PPE" is significant (P < 0.05) d The contrast "POO and POE vs PBO, PBE, PPO and PPE" is significant (P < 0.05) Table 31. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma iron and total iron binding capacity (TIBC) of colostrum deprived piglets (Experiment 4). Day 7 14 21 28 Treatment X S E X S E X S E X S E X S E X S E X SE Plasma Iron (ug dlr1) Control na na 298a 27 Non-est1 Non-est Non-est Non-est BBO na na 389 15 150b 20 134b 16 108 13 90 5 POO na na 401 15 215 15 194d 12 112d 10 94 4 POE na na 413 15 254c 15 190d 12 109d 10 86 4 PBO na na 402 15 182 16 154 12 109 11 84 4 PBE na na 434 15 223c 16 160 13 98 11 84 5 PPO na na 401 16 197 17 161 14 107 11 90 5 PPE na na 421 15 212c 15 170 12 110 10 92 4 TIBC (ug dlr1) Control 246 14 289 18 381a 30 Non-est Non-est Non-est Non-est BBO 233 13 294 17 502 19 461 21 487 20 544 18 579 12 POO 267 13 318d 17 493 19 461 16 520 15 545 14 577 10 POE 259 13 323d 17 494 19 495 16 502 15 549 14 566 10 PBO 246 13 263 17 494 19 469 17 496 16 567 15 572 10 PBE 253 13 275 17 526 19 450 17 491 16 550 15 585 10 PPO 242 13 296 17 508 20 485 18 508 17 577 15 586 11 PPE 250 13 299 17 503 19 455 17 502 15 537 14 574 10 1 Non-est = non-estimable a The contrast "Controls vs all other treatments" is significant (P < 0.05) b The contrast "BBO vs POO, POE, PBO, PBE, PPO and PPE" is significant (P < 0.05) c The contrast "POO, PBO and PPO vs POE, PBE and PPE" is significant (P < 0.05) d The contrast "POO and POE vs PBO, PBE, PPO and PPE" is significant (P < 0.05) 128 Table 32. The effect of EDDA, with and without bovine or porcine iimiunoglcibulins, on the packed cell volumes and hemoglobin concentrations of colostrum deprived piglets (Experiment 4). Day 0 1 4 7 14 21 28 Treatment X SE X SE X SE X SE X SE X SE X SE Packed Cell Volume (%) Control 32.7 1.7 30.2 1.6 33.5a 2.3 Non-est1 Non-est Non-est Non-est BBO 31.4 1.6 26.0 1.5 23.7 1.4 23.3 1.5 31.4 1.4 30.3 1.3 28.9 1.3 POO POE 34.3 31.9 1.6 1.6 28.9 1.5 27.4c 1.5 26.6 25.5 1.4 1.4 24.7 1.2 23.8c 1.2 31.4 1.1 26.1c 1.1 30.6 30.8 1.0 1.0 29.2 29.6 1.0 1.0 PBO PBE 32.6 32.9 1.6 1.6 30.6 1.5 27.9c 1.5 27.6 25.9 1.4 1.4 26.5 1.3 23.6c 1.3 31.1 1.2 28.2c 1.2 31.4 31.4 1.1 1.1 27.8 29.2 1.1 1.1 PPO PPE 33.6 31.7 1.6 1.6 30.4 1.5 26.0c 1.5 27.4 25.2 1.5 1.4 28.4 1.4 24.3c 1.2 33.3 1.2 27.3c 1.1 32.4 30.1 1.1 1.0 29.4 28.2 1.1 1.1 HQBGLOBIN (g dLr1) Control 13.1 0.9 11.7 0.7 13.3 2.1 Non-est Non-est Non-est Non-est BBO 12.0 0.8 10.3 0.6 9.6 1.3 9.3 0.6 12.6 0.7 11.4 0.6 11.0 1.6 POO POE 13.5 12.7 0.8 0.8 11.7 0.6 10.7c 0.6 10.2 10.2 1.3 1.3 9.1 0.5 9.1 0.5 10.9 0.6 10.1c 0.6 11.9 11.5 0.4 0.4 10.6 10.4 1.3 1.4 PBO PBE 13.4 13.2 0.8 0.8 12.2 0.6 11.0c 0.6 10.6 10.7 1.3 1.3 9.7 0.5 8.8 0.5 12.3 0.6 10.6c 0.6 11.9 11.6 0.5 0.5 11.1 11.2 1.4 1.4 PPO PPE 13.9 11.8 0.8 0.8 11.9 0.6 10.4c 0.6 10.0 9.9 1.4 1.3 10.4 0.5 9.0 0.5 12.7 0.6 10.8c 0.6 12.2 11.7 0.5 0.5 11.6 10.5 1.4 1.4 1 Non-est = non-estimable » The contrast "Controls vs all other treatments" is significant (P < 0.05) c The contrast "POO, PBO and PPO vs POE, PBE and PPE" is significant (P < 0.05) 129 Table 33. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma iinmunoglobulin concentrations of colostrum deprived piglets (Experiment 4). Treatment 0 1 4 7 day 14 21 2! 3 X SE X SE X SE X SE X SE X SE X SE Control <1.3 - <1.3a - <1.3a - Non-est1 Non-est Non-est Non-est BBO BIgG <0.5 - 5.7 0.1 5.2 0.1 3.5 0.1 1.9 0.1 0.7 0.1 <0.5 -PIgG <1.3 - <1.3 - <1.3 - 3.7 2.4 5.2 1.8 10.1 1.3 10.6 1.0 BBO Total IgG <1.3 - 5.7b 4.4 5.2b 3.7 7.2b 3.1 7.1b 2.3 10.8 1.5 10.6 1.2 POO <1.3 - 26.0 2.8 22.8 2.4 19.4 1.9 13.5 1.4 10.8 1.0 8.7 0.7 POE <1.3 - 24.6 2.9 22.5 2.5 17.9 2.0 12.7 1.5 9.9 1.0 8.7 0.8 PBO <1.3 - 29.5 2.9 27.6 2.5 23.8 2.0 15.3 1.5 12.2 1.0 10.1 0.8 PBE <1.3 - 27.2 2.9 25.0 2.5 18.0 2.1 12.4 1.5 10.5 1.0 9.7 0.8 PPO' <1.3 - 28.2 3.1 24.4 2.6 18.7 2.2 14.2 1.6 10.0 1.1 9.5 0.8 PPE <1.3 - 27.1 2.9 24.2 2.5 19.4 2.0 15.1 1.5 11.1 1.0 9.3 0.8 1 Non-est = non-estimable a The contrast "Controls vs all other treatments" is significant (P < 0.05) b The contrast "BBO vs POO, POE, PBO, PBE, PPO and PPE" is significant (P < 0.05) 130 Figure 16. The effect of EDDA, with and without bovine or porcine immunoglobulins, on the plasma immunoglobulin concentrations of colostrum deprived piglets (Experiment 4). For s t a t i s t i c a l analysis see Table 33. Day 131 Experiment 5 Introduction EDDA has been shown to have significant effects on iron metabolism and excretion in rats. In experiments 3 and 4 iron status was measured by hematocrit, hemoglobin, plasma iron and total iron binding capacity. These measures do not give a detailed picture of iron metabolism. They also do not give any information on the effect that EDDA has on the excretion of iron in the feces and urine. The use of radioactive isotopes of iron to study the metabolism of iron is a well established technique (Cavill, 1986). This objective of this experiment was to examine the effect that oral administration of EDDA had on iron metabolism and excretion in piglets using [ 5 9Fe]. Materials and Methods  Experimental Design • Piglets were randomly assigned to one of two treatments. A l l piglets received 1.5 L day - 1 of the same milk replacer used in Experiments 1,3 and 4. The Control group received milk replacer only. The EDDA group received 37.5 mg EDDA Kg body weight - 1 d a y 1 mixed with the milk replacer. The results were analyzed using the General Linear Models procedure of SAS. Animal Management Eight piglets that had completed Experiment 4 were used. They were randomly assigned to the two treatments. The piglets were fed their assigned diets for 3 days before the injection of [ S 9Fe]. Piglets were weighed daily and their EDDA intake was adjusted accordingly. The piglets were 28 days of age at the beginning of the experiment and weighed between 5,550 and 7,250 g. The piglets were housed individually in stainless steel metabolism cages measuring 1 m square. The cages allowed separate collection of urine and feces. 132 Preparation of Radioiron Labelled Plasma The plasma was labelled with [ 3 9Fe]Cl3 (Amersham Canada Ltd. Oakville Ont.) using the procedure described by Cook and Finch (1980). Ten mL of blood were collected from a pig in the same l i t t e r , but not in the same experiment. The blood was centrifuged and the plasma removed into a sterile tube. A solution of [ S 9Fe]Cl 3 (3-20 mCi mg"1 Fe) was prepared in .01 N HC1 with sufficient 3.8% sodium citrate to give a molar ratio in excess of 20:1. The solution was added dropwise to the plasma with constant sti r r i n g to give a f i n a l concentration of about 6 uCi mL-1 of plasma. The plasma was held at room temperature for at least 30 minutes before being injected into the piglets. Injection of Radioiron and Sampling regimen The anterior vena cava of each piglet was catheterized and the catheter was used for injection of [ 5 9Fe] and for taking a l l blood samples. Before injection, blood samples were taken for hematocrit and serum iron determin-ation. Standards were prepared from the labelled serum immediately before injecting the [ 5 9Fe]. The piglets were injected with about 1000 mg of the labelled plasma measured accurately to the nearest mg. The catheter was then flushed with st e r i l e saline to ensure that a l l of the labelled plasma entered the bloodstream. Blood samples were collected at 15,30,45,60 and 120 minutes after the injection to measure plasma iron disappearance and plasma iron tur-nover in the piglets. Samples were also taken at 24,72,120 and 144 hours to measure the incorporation of [ 3 9Fe] into the red blood c e l l s . Urine and feces were collected twice daily. After 6 days the piglets were sacrificed. Blood samples were taken and the livers and spleens were removed, weighed and samples taken for counting. Samples were counted using a Packard Auto-Gamma counter (Model 500c). 1 3 3 Calculations A l l calculations were made using formulas given by Cook and Finch (1980) . The injected [ S 9Fe] was calculated using the following formula: The plasma radioactivity at time 0 was calculated by taking the regression of the logarithm of plasma radioactivity (CPM) against time. The y intercept was used as the CPM at time 0. The Plasma Iron Disappearance Rate (T%) was calculated as the time required to. reduce the Time 0 plasma radioactivity by half. Plasma volume was calculated using the following formula: Plasma iron turnover (PIT) rate was calculated using the following formula: Standard [ S 9Fe] (CPM/mL) x weight of dose x 100 Injected [ 3 9Fe] (CPM) = Weight of standard Plasma volume .(ml) Injected [ 3 9Fe] (CPM) Time 0 plasma [3 9Fe] (CPM ml"1) Plasma iron (ug 100 mL-1) x (100 - Hematocrit) PIT (mgLblood-i day 1) = fJ4 x 100 Red c e l l u t i l i z a t i o n of [ 3 9Fe] was calculated as: Red c e l l [ 3 9Fe] (CPM mL"1) x red c e l l mass x 100 Red c e l l u t i l i z a t i o n (%) = Injected L a aFeJ activity where Plasma volume (mL) x 0.9 x Hematocrit Red c e l l mass (mL) = 100 - (0.9 x Hematocrit) 134 The distribution of [ s 9Fe] in urine, feces, liver and spleen was calculated as: Sample [ 3 9Fe] (CPM) x sample weight x 100 Sample distribution (%) = =—,—,—_—r-aw . _ — „ _ -Injected [ 5 9Fe] activity Results The only significant treatment differences occurred for VA and PIT (see Table 34). The Controls had a significantly lower T% than the EDDA group (75.5 vs 186.3 minutes). The Controls also had a significantly greater PIT than the EDDA group (0.90 vs 0.34 mg Kg - 1 day - 1). Figure 17 shows the plasma iron disap-pearance curve. There was an i n i t i a l rapid incorporation of [ 3 9Fe] into the red cells (See Figure 18) . After 24 hours a significantly higher percentage of the original dose of [ 3 9Fe] was incorporated into the red blood cells of the Controls. This difference remained significant throughout the experiment. [ 5 9Fe] incorporation reached a plateau on day 5 for the EDDA group but not for the Controls. The fecal and urinary excretion of [ 3 9Fe] was significantly higher for the EDDA treatment (See Figure 19). Excretion of [ 3 9Fe] in the feces by the EDDA group occurred at a f a i r l y constant rate throughout the 6 days o f the experi-ment. Urinary excretion occurred mainly in the f i r s t 24 hours after injection and then leveled off. Table 35 shows the excretion and distribution of [ 3 9Fe] at the end o f the experiment. The Controls excreted only 0.6% of the injected [ 3 9Fe] in the feces and 0.7% in the urine while the EDDA group excreted 4.0% in the feces and 6.3% in the urine. Both differences were significant (p > 0.05). The [ 5 9Fe] incorpor-ation into red blood cells was significantly lowered by EDDA. The amount o f [ 3 9Fe] in the spleen and liver was not affected by treatment. The residual 135 [ 5 9Fe] l e f t unaccounted for was not significantly different between the two treatments. The feces, urine, red blood cells, liver and spleen accounted for a l l but 1.9% of the injected [ 5 9Fe] in the Controls. The residual was 6.7% for the EDDA group. Discussion Each piglet consumed a l l the milk replacer i t was given. The average daily gains of piglets in the two treatments were not significantly different but there were large variations in piglet gains for the EDDA treatment. One piglet on this treatment gained only 33 g day - 1 while another gained 183 g day - 1. In Experiment 4, EDDA caused an overall decrease in average daily gains but here also the results were extremely variable. During the 14 day period in which EDDA was fed in Experiment 4, the average daily gains ranged from 42 to 130 g day - 1. In Experiment 3, during the same period gains ranged from 71 to 200 g day - 1. Only 2 piglets that received EDDA in Experiment 3 showed marked decreases in average daily gains. The method of feeding EDDA was different in the present experiment than in Experiments 3 and 4. The dose was 37.5 mg Kg - 1 in a l l 3 experiments but in Experiments 3 and 4 the concentration of EDDA in the milk replacer was a con-stant 0.1 mg ml - 1. In the present experiment the amount of milk replacer fed was a constant 1500 mL day - 1 and the concentration of EDDA varied with the weight of the piglet. This may have increased the amount of EDDA absorbed from the diet in the present experiment. EDDA had extremely variable effects on piglet performance. This could be due to differing a b i l i t i e s to absorb of EDDA from the diet or differences in the a b i l i t y of the piglets to detoxify the compound. EDDA is also toxic to sheep (Stifel and Vetter 1967). A dose of 50 mg mL-1 caused weight loss and death in 136 lambs. Plasma and red c e l l volumes were similar to those reported in other stu-dies. Talbot and Swenson (1970) found 4 week old piglets averaged about 75 mL Kg - 1 plasma volume and 25 mL Kg - 1 red c e l l volume. Jensen et al (1956) found growing pigs had plasma volumes ranging from 33.9-60.3 mL Kg - 1 and red c e l l volumes from 20-44.9 mL Kg - 1. Hematocrits were slightly lower than those expected for naturally reared piglets receiving 100 mg iron dextran injections. Talbot and Swenson (1970) found that 4 week old piglets that received 100 mg of iron dextran had hemato-crit s averaging 39.1. The values for piglets receiving no iron dextran averaged 20.8. The piglets used in this t r i a l received an iron deficient milk replacer as their sole source of iron. This probably caused the lowered hematocrits. Plasma iron was similar to those found in other studies (Furugouri 1971; Jensen et a l . 1956). Furugouri (1974) reported T% values ranging from 23 minutes for newborns to 30 minutes for 10 day old piglets. Studies with growing pigs found values ranging from 43-100 minutes (Jensen et a l . 1956; Furugouri et a l . 1974; Hristic et a l . 1970). The mean T& of 75.5 minutes for the Controls agrees well with these other studies. The EDDA piglets had a mean T% of 186.3 minutes. EDDA also affected plasma iron turnover (PIT). The Controls had a mean PIT of 0.90 mg Kg-1 day - 1. Other studies found values ranging from 0.4-2.0 mg Kg - 1 d a y 1 (Jensen et a l . 1956; Furugouri et a l . 1974; Hristic et a l . 1970). The EDDA treatment had a mean PIT of 0.34 mg Kg - 1 d a y 1 . EDDA has an apparently profound effect on the flow of iron in the plasma. This apparent effect may not be real. One possible interpretation of the results is that EDDA bound iron is a pool separate from transferrin bound iron. When 1 3 7 the injection of [ 3 9Fe] labelled transferrin enters the blood stream the [ 3 9Fe] is free to exchange with the EDDA pool. This means that [ 3 9Fe] remains in the bloodstream longer than i t would with no EDDA present. Analysis of plasma samples counts both the transferrin and EDDA pools of [ 3 9Fe]. This may lead to an apparent decrease in PIT where none exists. To test this hypothesis trans-ferrin and EDDA bound [ 3 9Fe] would have to be analyzed separately. During the f i r s t 24 hours after the injection of [ 3 9Fe], the Controls had a greater rate of [ 3 9Fe] incorporation into the red blood cells than piglets receiving EDDA. After 24 hours, however, the red blood c e l l [ 3 9Fe] incorporation curves were nearly parallel. The difference in [ 3 9Fe] incorporation between the Controls and the EDDA group is almost entirely explained by the increased excretion of iron in the feces and urine. Other studies reported red blood c e l l incorporation rates of from 72-100% (Jensen et a l . 1956; Braude et a l . 1962). The Controls f e l l within this range. EDDA caused an increase in the excretion of iron in the feces and urine. The only other studies of the effect of EDDA on iron excretion involved rats. An intramuscular injection of EDDA was used rather than the oral administration used in the present experiment. (Hershko et a l . 1984a; 1984b). EDDA caused sig-nificant excretion of iron in both urine and feces in these studies. Fecal ex-cretion was far greater than urinary excretion however. In the present study the reverse is true. Excretion in the feces depends on the [ 3 9Fe] g _ 1 of feces and the total amount of feces produced. The feces showed a high activity per gram but the total quantity of feces produced was small. The piglets, were fed only milk replacer. Since the di g e s t i b i l i t y of the milk replacer was nearly 100% this made for a small amount of fecal material and a lower excretion of [ 3 9Fe]. 1 3 8 In conclusion, EDDA has significant effects on iron metabolism. The most important is the increase in urinary and fecal excretion of iron. The increase in excretion leads to a decrease in the synthesis of red blood cells and hem-oglobin. This could be overcome by increasing the amount of supplemental iron dextran given to piglets. Of greater significance are EDDA's toxic effects. A significant number of piglets had reduced weight gains when EDDA was fed at 37.5 mg Kg body weight - 1. These toxic properties make i t unsuitable as an additive to piglet milk replacers. 139 Table 34. The effect of EDDA on individual piglets on the day of P'Fe] injection (Experiment 5). Average Plasma Red Cell Plasma Piglet Weight Gain Volume Volume Hematocrit Iron T< PIT (g) (g/dav) (mL) (mL) (%) (ua/dL) (min) (mg/kg/dav) Control Treatment 105 6,400 158 53.8 19.8 30.0 80 63.7 0.87 205 7,250 116 67.4 21.6 27.0 90 77.6 0.85 206 5,550 183 81.1 31.4 31.0 110 92.0 0.83 306 5,700 233 62.3 22.8 29.5 90 68.6 1.03 X 6,225 172 66.2 23.9 29.5 93 75.5a 0.90a SE 351 31 6.6 2.0 1.5 4.8 11.0 0.04 EDDA Treatment 102 6,400 75 57.5 23.8 32.5 90 198.2 0.31 201 6,650 166 58.3 28.6 36.5 100 215.4 0.29 203 6,750 183 48.7 24.3 37.0 95 183.3 0.33 303 5,400 33 82.8 27.4 29.0 90 148.2 0.43 X 6,300 114 61.8 26.0 33.8 94 186.3 0.34 SE 351 31 6.6 2.0 1.3 4.8 11.0 0.04 a The treatments means are significantly different (p > 0.05). 140 Table 35. The effect of EDDA on excretion and distribution of P'Fe],1 (Experiment 5). Feces Urine Red Cells Liver Spleen Residual Treatment X SE X SE X SE X SE X SE I SE Control 0.6a 0.8 0.7a 0.2 86.5a 1.8 8.0 2.0 2.3 0.3 1.9 3.2 EDDA 4.0 0.8 6.3 0.2 68.1 1.8 13.4 2.0 1.5 0.3 6.7 3.2 1 Values are in percent of injected ["Fe] activity. The values for feces and urine are from total collection during the entire experiment. The values for liver and spleen represent the total activity of the organ and includes the blood present within the organ. a The treatments means are significantly different (p > 0.05). 141 FIGURE 17. The e f f e c t of EDDA on plasma i r o n disappearance (Experiment 5 ) . M'nies 142 FIGURE 18. The effect of EDDA on the incorporation of [ 3 9Fe] into red blood cells (Experiment 5). Days 1 4 3 FIGURE 19. The effect of EDDA on urinary and fecal excretion of [ 5 9Fe] (Experiment 5). Feces 1 2 3 + 5 8 Day Urine 1 2 3 4 5 8 144 EXPERIMENT 6 Introduction In a l l previous experiments in this series, the concentration of plasma immunoglobulins was used as the only measure of piglet immunity. In this ex-periment i t was decided to examine the effect a r t i f i c i a l rearing on cell-me-diated immunity. Colostrum and milk contain viable leukocytes in addition to immunoglob-ulins. In mice and rats lymphocytes are absorbed from the milk into the blood-stream of the suckling animals (Parmely and Beer 1977; Head and Beer 1979). These cells can transfer c e l l mediated immunity from resistant mothers to suckling mice. Colostral lymphocytes may also be absorbed into the blood stream of the suckling piglet and contribute to i t s immune status. Colostrum deprived piglets do not receive viable lymphocytes and may have decreased c e l l mediated immunity compared to sow reared piglets. Ideally viable lymphocytes from sow colostrum should be fed to colostrum deprived piglets during the f i r s t day of l i f e . The population of lymphocytes found in colostrum is different than the one found in the peripheral blood (Parmely and Beer 1977) . A high proportion of colostral lymphocytes respond to enteric organisms and food antigens. Sow colostrum however, is d i f f i c u l t to obtain. For this reason i t was decided to investigate whether feeding peripheral blood lymphocytes to colostrum deprived piglets could increase c e l l mediated immunity. The colostrum deprived piglets were also compared to l i t t e r mates that remained with the sow. Materials and Methods  Experimental Design The treatments used are shown in Table 36. 145 Table 36. Experimental protocol used to study the effect of sow or a r t i f i c i a l rearing on colostrum deprived piglets (Experiment 6). Treatment dav 1 dav 2-14 dav 15-28 Sow reared Sow's Colostrum Sow's Milk Sow's Milk A r t i f i c i a l 25 mg mL-1 PIgG 5 mg mL-1 PIgG 0 mg mL-1 PIgG No leukocytes Leukocyte 25 mg mL-1 PIgG 5 mg mL-1 PIgG 0 mg mL-1 PIgG 5 x 108 leukocytes The piglets in this experiment were reared in four outcome groups. Outcome groups 1, 2 and 3 consisted of 6 piglets with 2 piglets assigned to the each treatment. Outcome group 4 consisted of 12 piglets with 4 piglets assigned to each treatment. Piglets that were not used in the experiment were l e f t with the sow. There was a minimum of 6 and a maximum of 8 piglets remaining with the sows for each outcome group. The outcome groups were a l l 1 week apart so that an " a l l - i n a l l out" regimen was not used. Piglets in a l l outcome groups started the experiment with other piglets already present in the room. The results were analyzed using the following least squares model. Yu = u + Ti + Gj + TiGj + Eij where Yu = the dependent variable u = the overall mean Ti = the effect of the i t h treatment Gj = the effect of the j t h outcome group 146 TiGj = the interaction between the i t h treatment and the j t h outcome group Eij = the residual error for each sample Non-significant interactions were added to the error term and the results recalculated. Treatment differences were measured using orthogonal contrasts. Leukocytes Leukocytes were separated from porcine abattoir blood. The method employed does not remove monocytes so the c e l l preparations are referred to as leuko-cytes. About 1 L of blood was collected from 1 pig at Intercontinental Packers Ltd. (Vancouver B.C.) using sodium EDTA as an anticoagulant. The blood was allowed to settle at room temperature for 2 hours. At this time the leukocytes formed a thick layer on top of the red blood c e l l s . The leukocytes were col-lected into a s t e r i l e flask using a water pump. The collected cells were mixed with 5 volumes of 0.83% ammonium chloride to lyse any contaminating red blood c e l l s . The leukocytes were harvested by centrifugation at 200 x g for 20 minutes and washed 3 times with Roswell Park Memorial Institute (RPMI) 1640 culture medium. The v i a b i l i t y .of cells isolated by this procedure were measured by Trypan Blue dye exclusion. The v i a b i l i t y was greater than 95% for a l l pre-parations. Individual doses of 5 x 108 cells were made up in 15 mL of RPMI-1640 containing 10% fetal calf serum, 50 IU mL-1 p e n i c i l l i n and 50 ug mL-1 streptomycin. The doses were kept at 4 °C until use. A l l doses were used within 24 hours of preparation. Doses were slowly warmed to 32 °C before being given to piglets. In vivo Cell-Mediated Immunity Cell-mediated immunity was measured as the intradermal response to the T c e l l mitogen phytohemagglutinin (PHA) (Blecha et a l . 1983). A solution (0.1 147 mL) containing 250 ug mL-1 PHA was injected into the medial aspect of one flank. The other flank was injected with s t e r i l e saline. Double skin fold thicknesses of each flank were measured at 24 and 48 hours after injection using a constant tension micrometer. The increase in thickness in the PHA flank minus the increase in thickness in the Control flank was used as the measure of in vivo cell-mediated immunity. Animal Management Piglets were removed from the sow at birth and placed in cages in the experimental room used in previous experiments. The piglets were then randomly assigned to treatments. The piglets on the leukocyte treatment received 5 x 108 leukocytes via intubation. The leukocytes were given in 15 mL RPMI-1640 culture media supplemented with 10% fetal calf serum, 50 IU mL-1 p e n i c i l l i n and 50 pg mL-1 streptomycin. The sow reared piglets and a r t i f i c i a l treatment piglets were intubated and given 15 mL RPMI-1640 containing 10% fetal calf serum, 50 IU mL-1 p e n i c i l l i n and 50 pg mL-1 streptomycin. The sow reared p i g l -ets were then returned to the sow where they remained for the duration of the experiment. The a r t i f i c i a l and leukocyte treatment piglets were started on PIgG fo r t i f i e d milk replacer and were reared identically to piglets in Experiment 4. A l l piglets were given 300,000 IU of p e n i c i l l i n on days 2, 4 and 6. This is a standard treatment for a l l sow reared piglets at UBC and i t was decided that the a r t i f i c i a l and leukocyte treatment piglets should receive the same treatment. Blood samples were taken from the suborbital sinus at birth and on days 1,7,14,21 and 28. The blood was analyzed for PIgG. Piglet weights were taken every.other day and the feed intake of the a r t i f i c i a l and leukocyte treatments 148 adjusted accordingly- Diarrhea scores were not kept in this experiment because of the d i f f i c u l t y of observing the feces of sow reared piglets. Results Survival One sow reared piglet was overlain by a sow (see Table 37). Aside from that there were no disease problems or diarrhea among the sow reared piglets. A l l of the a r t i f i c i a l and leukocyte treatment piglets survived the 28 day experimental period. There was a major disease problem among the last two outcome groups however. Two piglets on the a r t i f i c i a l treatment and one piglet on the leukocyte treatment were euthanized at the end of the experiment. These piglets were both from outcome group 4. They had severe diarrhea, weight loss and were unable to stand at the end of the experiment. Post-mortem examinations found the cause was septicemia due to Streptococcus suis II. Several piglets in outcome group 2 on the leukocyte treatment showed inflammation of the joints. These piglets did improve enough to be returned to the herd at the end of the experiment. Weight Gains and Piglet Weights Average daily gains were not significantly different for any treatment (see Table 38). There was a trend towards higher average daily gains for the sow reared piglets, especially during weeks 2 and 3. A l l treatments showed a depr-ession in average daily gains during week 4. Mean piglet weights for the sow reared piglets were significantly higher on days 14, 21 and 28 (see Table 39). At the end of the experiment the sow reared piglets had a mean weight of 7,360g compared to 6,310g for the leukocyte treatment piglets and 5,940 g for the a r t i f i c i a l treatment piglets. Plasma Immunoglobulin Concentrations 1 4 9 Tables 40 and Figure 20 show the plasma PIgG concentrations. Plasma PIgG of the sow reared piglets was significantly higher than the other treatments for a l l days. The sow reared piglets had a plasma PIgG level of 44.1 mg mL-1 on day 1 compared to 26.9 and 27.1 mg mL-1 for the a r t i f i c i a l and leukocyte treatments respectively. Intradermal Response to PHA The intradermal response to PHA is shown in Table 41. The sow reared piglets had a significantly greater increase in flank thickness at 24 and 48 hours than either of the other treatments. The interaction between treatment and outcome group was significant at 24 hours. Table 42 shows the treatment x outcome group means at 24 hours. There was a wide variation between outcome groups in the increase in flank thickness due to PHA. The sow reared piglets had a significantly greater response to PHA in outcome groups 3 and 4 only. The leukocyte treatment piglets had a significantly greater response to PHA than the a r t i f i c i a l treatment in outcome groups 1 and 4. There were no significant treatment differences at 24 hours for piglets in outcome group 2. Discussion One of the advantages of a r t i f i c i a l rearing is that i t eliminates traumatic injuries to the piglet by the sow. One sow reared piglet was crushed in this experiment. The crushing and trampling of piglets by the sow accounts for from 18-52% of a l l pre-weaning mortality (Braude et a l . 1954; Fraser 1966; Fahmy and Bernard 1971). Eliminating deaths due to trauma could save 1-2 weaned piglets per sow per year. A l l the a r t i f i c i a l and leukocyte treatment piglets survived. However, three piglets were too weak to return to the herd due to Streptococcus suis II 150 infection. Streptococcus suis II infection might have also been responsible for the decrease in weight gains for a l l treatments during week 4. The a r t i f i c i a l treatment in the present experiment and the PPO treatment in Experiment 4 were the same in a l l respects. The weight gains are similar for weeks 1, 2 and 3 in both experiments. During week 4 however, the piglets in the present experiment gained only 123.9 g day - 1. The piglets in Experiment 4 gained 209.5 g day - 1. The same is true of analogous treatments in Experiments 3 and 1. There was no decrease in average daily gains during week 4. A possible reason for this was a high level of environmental contamination during the present experiment. At one point a l l piglets on the a r t i f i c i a l and leukocyte treatments were present in the experimental room at the same time. This may have allowed a build up of environmental contagion greater than in previous experiments. The weights of the sow reared piglets were higher than those of the other two treatments from day 14 on. This could be due to several things. The feed intake of the a r t i f i c i a l and leukocyte treatment piglets was probably lower than the feed intake of the sow reared piglets (Braude et a l . 1970; Pettigrew et a l . 1985). Without actually measuring the feed intake of the sow reared pig-lets, i t is impossible to know how their feed intakes compared to the a r t i f i c i a l and leukocyte treatment piglets. Braude et a l . (1970) was able to increase the rate of gain of a r t i f i c i a l l y reared piglets by increasing the level of feeding. At a level of 100 g milk solids Kg body weight - 1 day - 1 piglets gained 326 g day-1 . In Experiments 1, 3, 4 and 6, piglets fed 75 mg milk solids Kg body weight- 1 day - 1 gained approximately 170-185 g day - 1 over the 28 day experimental period (See Table 43). Lewis et a l . (1978) examined the effect that milk yield had on piglet 151 weight gains. They found that sow milk yield and percent milk solids accounted for only 44% of the variation in piglet weight gains. Environmental, genetic and immunological factors also affect piglet gains. Since the piglets used were l i t t e r mates, environmental and immunological factors must be responsible for most of the remaining variation. There i s no way to quantify the effect of the differing environments on the piglets in the present experiment. Plasma PIgG and intradermal response to PHA partially measured the immunological d i f f e r -ences. The day 1 plasma IgG of the sow reared piglets averaged mean of 44.1 mg mL-1. This is similar to values of 40.2 mg mL-1 found by Klobassa et a l . (1981) and higher than the range of 18.7-39.0 mg mL-1 reported by Curtis and Bourne (1973). The plasma PIgG concentrations of the sow reared piglets were 14.4 mg mL-1 on day 28. This is very high compared to the 6.7 mg mL-1 found by Klobassa et a l . (1981) on the same day. The reason for this discrepancy may be due to the ELISA used to measure the plasma PIgG. The antibody to PIgG used in a l l of the experiments had speci f i c i t i e s for the heavy and light chains of the PIgG molecule. Since PIgM and PIgA have the same light chains as PIgG there is some cross reactivity present. The other studies cited measured PIgG, PIgM and PIgA and would have used antibodies that did not cross react with different isotypes. In spite of this, the comparison among treatments in this experiment were valid because the same assay was used for a l l measurements. The a r t i f i c i a l and leukocyte treatment piglets did not differ in plasma PIgG at any time during the experiment. Both treatments had plasma PIgG con-centrations similar to those found in piglets on analogous diets in Experiment 4. These values were significantly lower than for the sow reared piglets throughout the experiment. The higher level of PIgG in the serum of the sow 152 reared piglets may have contributed to their increased weight gains. The sow reared piglets received their immunoglobulins from colostrum and milk while the other two treatments received blood derived immunoglobulins. Sow colostrum contains primarily IgG (Porter and Chidlow 1979). The specific-i t i e s of colostral immunoglobulins are similar to those found in the blood. On day 1 therefore, both sow reared and a r t i f i c i a l l y reared piglets received immunoglobulins that were similar in nature. Sow milk contains mainly IgA and a high proportion of milk immunoglobulins are specific for enteropathogenic organisms encountered by the sow. On days 2-14, the sow reared piglets received mainly IgA that was highly specific for enteropathogenic organisms. The a r t i -f i c i a l l y reared piglets received immunoglobulins that were mainly IgG and unselected for speci f i c i t i e s against enteropathogens. In Experiments 3 and 4 i t was noted that piglets that received PIgG on day 1 a l l had comparable survival rates. The presence or absence of immuno-globulins in the diet on days 2-14 had no effect on piglet survival. One hypo-thesis i s that 20 mg mL-1 of blood derived PIgG on day 1 ensures the survival of colostrum deprived piglets. Dietary immunoglobulins after day 1 increase average daily gains. In the present experiment, the sow reared and a r t i f i c i a l l y reared piglets both received immunoglobulins of a similar type on day 1. Survival was the same for both groups. On days 2-28 the sow reared piglets received IgA highly specific for enteric organisms. The immunoglobulins fed to the a r t i f i c i a l l y reared piglets were the same as on day 1. Furthermore, immunoglobulins were only fed on days 2-14. The average daily gains of the sow reared piglets were sup-erior to the gains of the a r t i f i c i a l l y reared piglets. This effect may be due to the hypothesis given above. The day 1 immunoglobulins ensured the survival 153 of the piglets. After day 1, the immunoglobulins in sow milk provided superior protection from enteric infection and higher average daily gains. The intradermal response to PHA is a good measure of cell-mediated immunity in pigs, cattle and chickens (Blecha et a l . 1983; Haggard et a l . 1980; Regnier and Kelley 1981) . The sow reared piglets had a more prolonged response to PHA than the a r t i f i c i a l or leukocyte treatment piglets. At 48 hours they had a greater flank thickness than either the sow reared or leukocyte treatment piglets. The situation at 24 hours is less clear because of the interaction between outcome group and treatment. There was a large variation in the PHA response of the sow reared piglets in different outcome groups. For the sow reared piglets, this variation was correlated with the parity of the sows used in the experiment. The sow for outcome group 1 was a second parity sow and the increase in flank thickness at 24 hours was 2.2 mm. The sows for outcome groups 3, 4 and 5 were a l l fourth parity sows. They averaged 3.2, 4.7 and 4.7 mm. The level of immunity provided in the milk of the younger sow may have been less than for the older sows. There was a trend to lessened response to PHA from Outcome group 1 to Outcome group 4 in the a r t i f i c i a l and leukocyte treatments. Outcome groups 3 and 4 of these treatments had more disease problems than Outcome groups 1 and 2. Infection suppresses cell-mediated immunity (Beisel 1984). The lower PHA response in outcome groups 3 and 4 may have been caused by the increase in infection. As with immunoglobulins, the population of lymphocytes in milk is different from the population found in the bloodstream (Parmely and Beer 1977). A high proportion of them respond to enteric organisms and food antigens encountered by the mother. Feeding peripheral blood leukocytes would certainly have a 154 different effect than feeding lymphocytes found in milk. In the present experiment, there were significant improvements in response to PHA in leukocyte treatment piglets in outcome groups 1 and 4. There was no response in outcome groups 2 and 3 however. This may be due to the Major Histo-compatibility Complex (MHC) compatibility between the donor and recipient of the leukocytes. Normally a piglet would ingest lymphocytes from i t s mother. The piglets own lymphocytes would share at least half of the MHC antigens expressed by these ingested lymphocytes. The leukocyte treatment piglets received lym-phocytes from an unrelated donor. The donor and recipient may have none or many MHC antigens in common. Many functions of c e l l mediated immunity are MHC restricted. The closeness of the match between MHC types of the donor and recipient may determine i f intradermal response to PHA is increased or not. This has some relevance to the practice of cross fostering piglets at birth. Head and Beer (1979) found that rats that suckled their mothers before being fostered to a MHC incompatible mother developed normal immune responses. Rats moved to MHC incompatible mothers at birth showed decreased responsiveness to alloantigens and 38% died of graft-versus host disease. A similar experiment using pigs would be of considerable interest. In conclusion, sow reared piglets gained weight faster, had higher levels of plasma PIgG and 2 of 4 l i t t e r s , had greater intradermal responses to PHA compared to the a r t i f i c i a l treatment piglets. Feeding peripheral blood leuk-ocytes in addition to immunoglobulins caused significant increases in intra-dermal PHA responses in 2 of 4 l i t t e r s but this did not have any effect on average daily gains. Incompatibility between the MHC of the donor.leukocytes and the piglets that received them may be responsible for the inconsistent results. 1 5 5 Table 37. The effect of sow or artificial rearing on piglet survival (Experiment 6). day Birth 7 14 21 28  Treatment N N N N N % SE Sow Reared 10 9 9 9 91 91 6 Artificial 10 10 10 10 10 100 6 Leukocyte 10 10 10 10 10 100 6 1 There were no significant contrasts between treatment means (P < 0.05). 156 Table 38. The effect of sow or artificial rearing on piglet average daily gains (Experiment 6). Average Daily Gains (g dayl) week 1 week 2 week 3 week 4 Treatment X SE X SE X SE X SE Sow Reared 158.41 18.2 260.4 23.6 260.4 23.0 191.9 24.2 Artificial 140.3 19.6 196.0 25.4 207.8 24.7 123.9 11.6 Leukocyte 132.8 18.2 211.8 23.6 219.8 23.0 155.8 9.4  1 There were no significant contrasts between treatment means (P < 0.05). Table 39. The effect of sow or artificial rearing on mean piglet weights (g) (Experiment 6), Birth dav 7 dav 14 dav 21 day 28 Treatment X SE X SE X SE X SE X SE Sow Reared 1,240 46 2,370 127 4,200a 222 6,020a 310 7,360 434 Artificial 1,340 49 2,250 137 3,620 239 5,070 334 5,940 468 Leukocyte 1,330 47 2.190 128 3,680 223 5,210 311 6,310 435 a The contrast "Sow reared vs Artificial and Leukocyte"is significant (P < 0.05). Table 40. The effect of sow or artificial rearing on plasma PIgG levels (mg mL-1) (Experiment 6). day 0 _1 7 14 21 28 Treatment X S E X S E X S E X S E X S E X S E Sow Reared <0.1 - 44.1a 1.8 36.4a 1.6 27.2a 1.6 18.9a 1.3 14.4a 0.7 Artificial <0.1 - 26.8 1.9 19.8 1.7 13.0 1.6 9.3 1.2 8.8 0.7 Leukocyte <0.1 - 27.0 1.8 20.8 1.6 15.2 1.5 9.9 1.2 8.4 0.6 a The contrast "Sow reared vs Artificial and Leukocyte"is significant (P < 0.05). 158 Table 41. The effect of sow or artificial rearing on piglet intradermal response to PHA at 3 weeks of age (Experiment 6). Increase in Flank Thickness1 Treatment 24 Hrs 48 Hrs X SE X SE Sow Reared 3.7a 0.3 2.8a 0.3 Artificial 2.2 0.3 1.6 0.3 Leukocyte 2.5 0.3 1.5 0.3 1 Increase in Flank thickness =(PHA flank thickness at 24 or 48 hours - preinjection thickness) - (Saline flank thickness at 24 or 48 hours - preinjection thickness). a The contrast "Sow reared vs Artificial and Leukocyte"is significant (P < 0.05). 159 Table 42. The effect of sow or artificial rearing on piglet intradermal response1 to PHA at 24 hours for treatment x outcome group interaction (Experiment 6). Outcome group 1 2 3 4 Treatment X SE X SE X SE X SE Sow Reared 2.2 0.3 3.2 0.3 4.7a 0.3 4.7a 0.3 Artificial 2.5b 0.3 2.9 0.4 2.1 0.3 1.5b 0.2 Leukocyte 3.2 0.3 2.9 0.3 2.0 0.3 2.1 0.2 1 Increase in Flank thickness =(PHA flank thickness at 24 hours - preinjection thickness) - (Saline flank thickness at 24 hours - preinjection thickness). a The contrast "Sow reared vs Artificial and Leukocyte"is significant (P < 0.05). b The contrast "Artificial vs Leukocyte" is significant (P < 0.05). 160 FIGURE 20 . The effect of sow or a r t i f i c i a l rearing on piglet plasma PIgG (Experiment 6). For s t a t i s t i c a l analysis see Table 40. Day 161 GENERAL DISCUSSION Piglets require porcine immunoglobulins on day 1 for adequate passive systemic immunity. Piglets fed porcine immunoglobulins had plasma IgG levels many times higher than piglets fed bovine immunoglobulins. They also had in-creased survival rates and higher average daily gains (See Table 43). A f t e r day 1, the addition of immunoglobulins to the diet results in higher average daily gains but does not increase survival (See Table 43). In Experiment 4, i t was shown that bovine and porcine immunoglobulins are equally effective at increasing average daily gains when fed on days 2-14. The important consider-ation w i l l be the relative costs of porcine and bovine immunoglobulins. HBED has no potential as an additive to milk replacers for a r t i f i c i a l l y reared piglets. It had no in vitro anti-E^ c o l i activity. When added to piglet diets i t lower survival rates and decreased average daily gains. EDDA inhibited the growth of E^ c o l i in an in vitro test. When added to piglet diets on days 2-14 in Experiment 3, i t increased piglet average daily gains as much as porcine immunoglobulin. However, when added to diets on day 1-14 in Experiment 4, i t caused a decrease in average daily gains (See Table 43). This may be due to increased absorption of EDDA on day 1 before gut closure takes place. EDDA increased the excretion of iron in the urine and feces thus reducing the iron available for the synthesis of hemoglobin. Packed c e l l volumes and hemoglobin concentrations were low in piglets fed EDDA in Experiment 4. Since EDDA was not superior to porcine immunoglobulins when fed on days 2-14 EDDA is not re-commended as an additive to piglet milk replacers. The c e l l mediated immunity of a r t i f i c i a l l y reared piglets was impaired compared to sow reared piglets in 2 of 4 l i t t e r s . The feeding of leukocytes to a r t i f i c i a l l y reared piglets significantly increased c e l l mediated immunity in 162 2 of 4 l i t t e r s . The major histocompatibility complex types of the donor and recipients of the leukocytes may be responsible for the uneven results. Future experiments should measure the MHC types of the pigs involved. In addition to the uneven results, increases in the intradermal responses to PHA were not correlated with increases in average daily gain. Clearly, more work needs to be done in this area. The questions that need to be addressed are: 1) the normal levels of c e l l mediated immunity in piglets; 2) the extent to which c e l l mediated immunity is transferred from sow to piglet; 3) the effect of c e l l mediated immunity on piglet survival and average daily gains and 4) the effect of MHC compatability in the transfer of c e l l mediated immunity via dietary leukocytes. The survival of the piglets that received porcine immunoglobulins on day 1 and excluding piglets that received HBED was 116 out of 121 or 95%. Only piglets that were free from defects and weighed over 800 g were used so the results are not comparable to survival in commercial herds. Ten naturally reared piglets in Experiment 6 had a survival rate of 9 out of 10 piglets. These piglets were chosen using the same c r i t e r i a as for the a r t i f i c i a l l y reared piglets. The one sow reared piglet that died in Experiment 6 was crushed by the sow. The major advantage of a r t i f i c i a l rearing is the reduction of traumatic injuries caused by the sow. This factor alone accounts for up to one half of pre-weaning mortality (Fahmy and Bernard 1971). While there are many factors in sow's milk and colostrum that protect piglets against enteric infections, immunoglobulins alone provide an adequate level of protection. The a r t i f i c i a l rearing of piglets using milk replacers f o r t i f i e d with 25 mg mL-1 PIgG on day 1 and 5 mg mL-1 of either PIgG or BIgG on days 2-14 is a viable alternative to natural rearing. 163 Table 43. The average daily gains and survival of piglets in Experiments 1, 3, 4 and 6. Experiment n Ig day 1 Ig day 2 Chelator Survival Mean Daily Gain1 (mg/mL) i [mg/mL) (mg/mL) (%) (a/d; 1 11 0 0 0 19 66 4 9 0 0 0 0 -1 11 20 BIgG 4 BIgG 0 72 156 4 10 25 BIgG 5 BIgG 0 60 128 4 10 25 PIgG 5 BIgG 0 90 184 4 10 25 PIgG 5 BIgG 0.1 EDDA 93 151 1 12 20 PIgG 4 PIgG 0 92 169 3 8 20 PIgG 0 PIgG 0 88 159 3 7 20 PIgG 4 PIgG 0 100 178 3 7 20 PIgG 0 PIgG 0.1 EDDA 100 179 3 7 20 PIgG 4 PIgG 0.1 EDDA 100 174 3 8 20 PIgG 0 PIgG 0.1 HBED 62 136 3 7 20 PIgG 4 PIgG 0.1 HBED 70 170 4 10 25 PIgG 0 PIgG 0 100 150 4 10 25 PIgG 0 PIgG 0.1 EDDA 100 149 4 10 25 PIgG 5 PIgG 0 88 181 4 10 25 PIgG 5 PIgG 0.1 EDDA 100 154 6 10 25 PIgG 5 PIgG 0 100 164 6 10 25 PIgG 5 PIgG 0 100 178 6 10 Sow Colostrum and Milk 91 219 1 Mean daily gain is for the entire 28 days of the experiments. 164 CONCLUSIONS 1) Bovine serum immunoglobulins are poorly absorbed from the diet into the blood stream in the f i r s t 24 hours after birth and porcine serum immunoglobulins are required during this period in order to ensure adequate passive systemic immunity. 2) Bovine and porcine immunoglobulins are equally effective at increasing average daily gains on days 2-14. 3) The c e l l mediated immunity of a r t i f i c i a l l y reared piglets may be impaired compared to sow reared piglets. The feeding of leukocytes to a r t i f i c i a l l y reared piglets significantly increased c e l l mediated immunity in 2 out of 4 l i t t e r s . It had no effect on survival or average daily gains however. 4) The best average daily gains of any treatment group of a r t i f i c i a l l y reared piglets in these experiments (184 g d - 1 ) , were unacceptably low compared to those of sow reared piglets. Further studies on the ideal plane of feeding are required. 5) HBED has no potential as an additive to milk replacers for a r t i f i c i a l l y reared piglets. 6) EDDA is not recommended as an additive to piglet milk replacers due to i t s potential toxicity, i t s effects on iron metabolism and the fact that i t is not demonstrably superior to immunoglobulins. 165 LITERATURE CITED Aasa R. Malmstrom B.G. Saltman P. and Vanngard J.. 1963. The specific binding of iron III and copper II to transferrin and conalbumin. Biochim. Biophys. Acta 75:203-222. Aisen P. Leibman A. and Zweier J.. 1978. Stoichiometric and site characteristics of the binding of iron to human transferrin. J. Biol. Chem. 253:1930-1937. Allen W.D. and Porter P.. 1977. The relative frequencies and distribution of immunoglobulin-bearing cells in the intestinal mucosa of neonatal and weaned pigs and their significance in the development of secretory immunity. Immunol. 32:819-824. Asherson G.L. Colizzi V. and Zembala M.. 1986. An overview of T-suppressor c e l l c i r cuits. Ann. Rev. Immunol. 4:37-68. Aumaitre A. and Seve B.. 1978. Nutritional importance of colostrum in the piglet. Ann. Rech. Vet. 9:181-192. Babbitt B.P. Allen P.M. Matsueda G. Haber E. and Unanue E.R.. 1985. Binding of immunogenic peptides to la histocompatibility molecules. Nature 317:359-361. Baer A. Oroz M. and Blanc B.. 1979. Serological and fluorescence studies of the heat s t a b i l i t y of bovine lactoferrin. J. Dairy Res. 46:83-93. Banks K.L.. 1981. Host defense in the newborn animal. J. Amer. Vet. Med. Assoc. 181:1053-1056. Bates G.W. and Schlabach M.R.. 1973. A study of the anion binding site of trans-ferrin. FEBS Lett. 33:289-292. Beisel W.R.. 1984. Nutrition, infection, specific immune responses, and non-specific host defenses: A complex interaction, in. Nutrition, disease resis-tance, and immune function, ed. Watson R.R.. pp. 3-34. Marcel Dekker: New York. Beresford. C.H. Neal R.J. and Brooks O.G.. 1971. Iron absorption and pyrexia. Lancet i:568-572. Betts A.O. Lamont P.H. and Littlewort M.C.G.. 1960. The production by hyster-ectomy of pathogen free colostrum deprived pigs and the foundation of a minimal disease herd. Vet. Rec. 72:461-468. Bezkorovainy A.. 1980. Chemistry and metabolism of the transferrins, in. Bio-chemistry of nonheme iron. ed. Bezkorovainy A., pp. 127-206. Plenum Press: New York. Blecha F. Pollmann D.S. and Nichols D.A.. 1983. Weaning pigs at an early age decreases cellular immunity. J. Anim. Sci. 56:396-400. 166 Bockman D.E. and Cooper M.D.. 1973. Pinocytosis by epithelium associated with lymphoid f o l l i c l e s in the bursa of Fabricius, appendix and Peyer's patches. An electron microscopic study. Amer. J. Anat. 136:455-478. Bohl E.H. Gupta R.K.P. Fernando-Olquin M.V. and Saif L.J.. 1972. Antibody responses in serum, colostrum and milk of swine after infection or vaccination with transmissible gastroenteritis virus. Infect. Immun. 6:289-301. Borsos T. Chapuis R.M. and Langone J.J.. 1981. Distinction between fixation of CI and activation of complement by natural IgM anti-Hapten antibody: Effects of c e l l surface hapten density. Mol. Immunol. 18:863-868. Borsos T. and Rapp H.J.. 1965. Complement fixation on c e l l surfaces by 19S and 7S antibodies. Science 150:505-506. Bourne F.J. and Curtis J.. 1973. The transfer of immunoglobulins IgG, IgA and IgM from serum to colostrum and milk in the sow. Immunol. 24:157-162. Bourne F.J.. 1973. The immunoglobulin system of the suckling pig. Proc. Nutr. Soc. 32:205-215. Bradley P.A. Bourne F.J. and Brown P.J.. 1976. The respiratory tract immune system in the pig. 1. Distribution of immunoglobulin containing cells in the respiratory tract mucosa. Vet. Pathol. 13:81-88. Brandon M.R. Watson D.L. and Lascelles A.K.. 1971. The mechanism of transfer of immunoglobulin into mammary secretion of cows. Aust. J. Biol. Med. Sci. 49:613-623. Braude R. Mitchell K.G. Newport M.J. and Porter J.W.G.. 1970. A r t i f i c i a l rearing in pigs. I. Effect of frequency and level of feeding on performance and d i -gestion of milk proteins. Brit. J. Nutr. 24:501-516. Braude R. Kon S.K and Thompson S.Y.. 1947. A study of the composition of sow's milk. Brit. J. Nutr. 1:64-72. Braude R. Chamberlain M. Kotarbinska M. and Mitchell K.G.. 1962. The metabolism of iron in piglets given labelled iron either orally or by injection. Brit. J. Nutr. 16:427-449. Braude R. Clarke P.M. and Mitchell K.G.. 1954. Analysis of the breeding records of a herd of pigs. J. Agric. Sci. Camb. 45:19-27 Brock J.H. Arzabe F. Lampreave F. and Pineiro A.. 1976. The effect of trypsin on bovine transferrin and lactoferrin. Biochim. Biophys. Acta 446:214-225. Brock J.H.. 1985. Transferrins, in. Metalloproteins Part 2: Metal proteins with non-redox roles, ed. Harrison P.M.. pp.183-262. Verlag Chemie: Weinheim. Brown E.J. Joiner K.A. and Frank M.M.. 1984 Complement, in. Fundamental Immunology, ed. W.E. Paul. pp. 645-668. Raven Press: New York. 167 Brown P.J. Bourne F.J. and Denny M.R.. 1975. Immunoglobulin-containing cells in pig mammary gland. J. Anat. 120:329-335. Bullen J.J. Rogers H.J. and Griffiths E.. 1978. Role of iron in bacterial in-fection. Curr. Top. Microbiol. Immunol. 80:1-35. Bullen J.J. Rogers H.J. and Leigh L.. 1972. Iron-binding proteins in milk and resistance to Escherichia c o l i infection in infants. Brit. Med. J. 1972, 1 69-65. Bustad L.K. Ham W.E. and Cunha T.J..1947. Preliminary observations on using a synthetic milk for raising pigs from birth. Arch. Biochem. 17:249-256. Carlsson J.. 1980. Bactericidal effect of hydrogen peroxide is promoted by the lactoperoxidase-thiocyanate system under anaerobic conditions. Infect Immun. 29:1190-1192. Cartwright G.E. Lauritren M.A. Humphreys S. Jones J. Merrill T.M. and Wintrobe M.M. 1946. The anemia of infection. II. The experimental production of hypo-ferremia and anemia in dogs. J. Clin. Invest. 25:81-88. Castellino F.J. Fish W.W. and Mann K.G.. 1970. Structural studies on bovine lactoferrin. J. Biol. Chem. 245:4269-4275. Cavill I.. 1986. Plasma clearance studies, in. Radionuclides in Haematology, eds. Lewis S.M. and Bayly R.J., pp. 214-244. Churchill Livingstone: Edinburgh. Cepica A. and Derbyshire J.B.. 1984a. The effect of adoptive transfer of mononuclear leukocytes from an adult donor on spontaneous cell-mediated cytotoxicity and resistance to transmissible gastroenteritis in neonatal piglets. Can. J. Comp. Med. 48:360-364. Cepica A. and Derbyshire J.B.. 1984b. Antibody-dependent and spontaneous c e l l -mediated cytotoxicity against transmissible gastroenteritis virus infected cells by lymphocytes from sows, fetuses and neonatal piglets. Can. J. Comp. Med. 48:258-261. Chapman H.A. Johnson J.S. and Cooper M.D.. 1974. Ontogeny of peyer's patches and immunoglobulin containing cells in pigs. J. Immunol. 112:555-563. Chu R.M. Glock R.D. Ross R.F. and Cox D.F.. 1979. Lymphoid tissues of the small intestine of swine from birth to one month of age. Amer. J. Vet. Res. 40:1713-1728. Cook J.D. and Finch C.A.. 1980. Ferrokinetic measurements, in. Iron. ed. Cook J.D. pp. 134-147. Churchill Livingstone: Edinburgh. Cooper M.D. Kearney J. and Scher I.. 1984. B lymphocytes, in. Fundamental Immunology, ed. Paul W.E.. pp.43-55. Raven Press: New York. 168 Cooper M.D.. 1981. Pre-B c e l l s : Normal and abnormal development. J. Clin. Immunol. 1:81-90. Curtis J. and Bourne G.J.. 1973. Half lives of immunoglobulins IgG, IgA and IgM in the serum of newborn pigs. Immunol. 24:147-155. Day N.K.B. Pickering R.J. Gewurz H. and Good R.A.. 1969. Ontogenic development of the complement system. Immunol. 16:319-326. Dembic Z. Haas W. Weiss S. McCubrey J. Kiefer H. von Boehmer H. and Steinmetz M. 1986.. Transfer of specificity by murine a and fi T-cell receptor genes. Nature 320:232-238. Dobrescu L. and Huygelen C . 1976. Protection of piglets against neonatal E.  co l i enteritis by immunization of the sow with a vaccine containing heat-labile enterotoxin (LT). Zbl. Vet. Med. B 23:79-88 E l l i o t J.I. Senft B. Erhardt G. and Fraser D.. 1984. Isolation of lactoferrin and i t s concentration in sow's colostrum and milk during a 21-day lactation. J. Anim. Sci. 59:1080-1084. E l l i o t J.I. Modler H.W. Timbers G.E. and Allen R.W.. 1987. A continuous process for the production of purified porcine gammaglobulin for use in pig milk re-placer supplements. Anim. Feed Sci. Tech. 17:213-218. Emmrich F. Strittmatter V. and Eichmann K.. 1986. Synergism in the activation of human CD8 T cells by crosslinking the T c e l l receptor complex with the CD8 differentiation antigen. Proc Natl. Acad. Sci. USA 83:8298-8302. Evans P.A. Newby T.J. Stokes C.R. and Bourne F.J.. 1982. A study of cells in the mammary secretions of sows. Vet. Immunol. Immunopathol. 3:515-527. Fahmy M.H. and Bernard C . 1971. Causes of mortality in Yorkshire pigs from birth to 20 weeks of age. Can. J. Anim. Sci. 51:351-359 Fielding J. 1980. Serum iron and iron binding capacity, in . Iron. ed. Cook J.D.. pp. 15-43. Churchill Livingstone: New York. Fisher M.M. Nagy B. Bazin H. and Underdown B.J.. 1979. Biliary transport of IgA. Role of secretory component. Proc. Natl. Acad. Sci. USA 76:2008-2012. Ford J.E.. 1974. Some observations on the possible nutritional significance of vitamin B12 and folate binding proteins in milk. Brit. J. Nutr. 31:243-257. Ford-Hutchinson A.W. and Perkins D.J.. 1972. Chemical modifications of the tryptophanyl groups of transferrin. Eur. J. Biochem. 25:415-419. Franek M. Prochazka Z. Franz J. Krejci J. and Mensik J.. 1975. The transfer of 1 3 1 I - l a b e l l e d immunoglobulins from serum to colostrum and milk in the sow. Acta. Vet. Brno. 44:93-103. 169 Fransson G-B. Toren-Tolling K. Jones B. Hambraeus L. and Lonnerdal B.. 1983. Absorption of lactoferrin-iron in suckling pigs. Nutr. Res. 3:385-394. Franz J. Milon A. and Salmon H.. 1982. Synthesis of immunoglobulins IgG, IgM and IgA during the ontogeny of foetal pigs. Acta. Vet. Brno 51:23-30. Fraser A.F.. 1966. Studies of piglet husbandry in Jamaica. II: Principal causes of loss between birth and weaning. Brit. Vet. J. 122:325-331 Frost A.E. Freedman H.H. Westerbach S.J. and Martell A.E.. 1958. Chelating tendencies of N,N'-Ethylenebis-[2-(o-hydroxyphenyl)]-glycine. J. Amer. Chem. Soc. 80:530-536. Fu S.M. Winchester R.J. and Kunkel H.G.. 1975. Similar idiotypic specificity for the membrane IgD and IgM of human B lymphocytes. J. Immunol. 114:250-252. Furugouri K. 1971. Normal values and physiological variations of plasma iron and total iron-binding capacity in pigs. J. Anim. Sci. 32:667-672. Furugouri K. 1972. Plasma iron and total iron-binding capacity in piglets in anemia and iron administration. J. Anim. Sci. 34:421-426. Furugouri K. 1973. Developmental changes in the nonheme iron composition of the liver and spleen in piglets. J. Anim. Sci. 36:265-270. Furugouri K. 1974. Kinetics of iron metabolism in piglets. J. Anim. Sci. 38:1249-1256. Genco R.J. Linfer R. and Evans R.T.. 1983. Effect of adjuvants on orally administered antigens. Ann. NY Acad. Sci. 409:650-667. Giovaniello T.J. DiBenedetto G. Palmer D.N. and Peters T. Jr. 1968. Fully auto-mated method for the determination of serum iron and total iron-binding cap-acity, in. Automation in analytical chemistry, Technicon Symposia 1967, Vol. 1. pp. 185-188. Mediad Inc.: White Plains, N.Y.. Goldblum R.M. Ahlstedt S. Carlsson B. Hanson L.A. Jodal U. Lidin-Janson G. and Sohl-Akerlund A.. 1975. Antibody-forming cells in human colostrum after oral immunization. Nature 257:797-799. Goldsmith S.J. Dickson J.S. Barnhart H.M. Toledo R.T. and Eitenmiller R.R.. 1983. IgA, IgG, IgM and lactoferrin contents of human milk during early lact-ation and the effect of processing and storage. J. Food Prot. 46:4-7. Grady R.W. and Jacobs A.. 1981. The screening of potential iron chelating drugs, in. Development of iron chelators for c l i n i c a l use. eds. Martell A.E. Anderson W.F. and Badman D.G.. pp. 133-164. Elsevier-NDU: Amsterdam. Gregory M.E. and Holdsworth E.S.. 1955. The occurance of a cyanocobalamin-binding protein in milk and the isolation of a cyanocobalamin-protein complex from sows milk. Biochem J. 59:329-334. 170 Grifiss J.M.. 1975. Bactericidal activity of meningococcal antisera. Blocking by IgA of l y t i c antibody in human convalescent sera. J. Immunol. 114:1179-1183. Groves M.L.. 1960. The isolation of a red protein from milk. J. Amer. Chem. Soc. 82:3345-3350. Haggard D.L. Stowe H.D. Conner G.H. and Johnson D.W.. 1980. Immunologic effects of experimental iodine toxicosis in young cattle. Amer. J. Vet. Res. 41:539-543. Haye S.N. and Kornegay E.T.. 1979. Immunoglobulin G, A and M and antibody response in sow-reared and arti f i c i a l l y - r e a r e d pigs. J. Anim. Sci. 48:1116-1122. Head J.R. and Beer A.E.. 1979. In vivo and in vitro assessment of the immuno-logic role of leukocytic cells in milk. in. Immunology of breast milk. eds. Ogra P.L. and Dayton D.H.. pp. 207-221. Raven Press: New York. Hemmings W.A.. 1980. First experience of dietary antigen. Lancet 1:818. Henry C. and Jerne N.K.. 1968. Competition of 19s and 7s antigen receptors in the regulation of primary immune response. J. Exp. Med. 128:133-152. Herberman R.B. and Ortaldo J.R.. 1981. Natural k i l l e r c e l l s : Their role in defence against disease. Science 214:24-32. Hershko C. Grady R.W. and Link G. 1984a. Development and evaluation of the im-proved iron chelating agents EHPG, HBED and their dimethyl esters. Haematologia 17:25-33. Hershko C. Grady R.W. and Link G. 1984b. Phenolic ethylenediamine derivatives: A study of orally effective iron chelators. J. Lab. Clin. Med. 98:337-347. Holmgren J. Svennerholm A.M. and Linblad M.. 1983 . Receptor-like glycocompounds in human milk that inhibit classical and El Tor Vibrio cholerae c e l l adherence (hemagglutination). Infect. Immun. 39:147-154. Hoover R.G. and Lynch R.G.. 1983. Isotype-specific suppression of IgA: Suppr-ession of IgA responses in BALB/c mice by T a ce l l s . J. Immunol. 130:521-523. Hristic V. Nikolic A. and Stovsic D.. 1970. Some effects of protein deficiency in young growing pigs. III. Ferrokinetic investigations. Acta Vet. Scand. 11:16-21. Ishizaka K. Ishizaka T. Lee E.H. and Fudenberg H.. 1965. Immunochemical pro-perties of human xA isohemagglutinin. I. Comparisons with xG- and xM-globulin antibodies. J. Immunol. 95:197-208. Jenness. R.. 1981. Inter-species comparison of milk proteins. Dev. Dairy Chem. 1:87-114. 171 Jensen W.N. Bush J.A. Ashenbrucker H. Cartwright G.E. and Wintrobe M.M.. 1956. The kinetics of iron metabolism in normal growing swine. J. Exp. Med. 103:145-159. Johanssen B.G.. 1969. Isolation of crystalline lactoferrin from human milk. Acta Chem. Scand. 23:683-684. Jonnson A.. 1973. Transfer of immunoglobulins from mother to offspring in the pig. Acta Vet. Scand. 43:Suppl. 7-64. Kawanishi H. Saltzman L.E. and Strober W.. 1983. Mechanisms regulating IgA class-specific immunoglobulin production in murine gut-associated lymphoid tissues. I. T cells that switch slgM B cells to slgA B cells in vitro. J. Exp. Med. 157:433-450. Kay R.M. Gleed P.T. Patterson A. and Sansom B.F. 1980. Effects of low level dosing of iron on the haematology and growth rate of piglets. Vet. Rec. 106:408-410. Kazatchkine M.D. Fearon D.T. and Austen K.F.. 1979. Human alternative complement pathway: Membrane-associated s i a l i c acid regulates the competition between fi and filH for c e l l bound C3b J. Immunol. 122:75-81. Kende M. 1982. Role of macrophages in the expression of immune responses. J . Amer. Vet. Med. Assoc. 181:1037-1042. Kincaid P.W.. 1981. Formation of B lymphocytes in fetal and adult l i f e . Adv. Immunol. 31:177-245. Kiyono H. Mosteller-Varnum L.M. Pitts A.M. Williamson S.I. Michalek S.M. and McGhee J.R.. 1985. Isotype-specific immunoregulation. IgA binding factors produced by Fc a receptor-positive T c e l l hybridomas regulate IgA responses. J . Exp. Med. 161:731-747. Klebanoff S.J. and Clark R.A.. 1978. The neutrophil. Elsevier:New York Klobassa F. Werhahn E. and Butler J.E.. 1981. Regulation of humoral immunity in the piglet by immunoglobulin of maternal origin. Res. Vet. Sci. 31:195-206. Kluger M.J. and Rothenberg B.A.. 1979. Fever and reduced iron: Their interaction as a host defense response to bacterial infection. Science 203:374-376. Knight CD. Klasing K.C. and Forsyth D.M. 1984. The effects of intestinal Escherichia c o l i 263, intravenous infusion of Escherichia c o l i 263 culture f i l t r a t e and iron dextran supplementation on iron metabolism in the young pig. J. Anim. Sci. 59:1519-1528. Kochan I.. 1977. Role of siderophores in nutritional immunity and bacterial parasitism, in. Microorganisms and minerals, ed. Weinberg E.D.. pp. 273-298. New York: Marcel Dekker. 172 Korsmeyer S.J. Hieter P.A. Ravetch J.V. Poplack D.G. Waldmann T.A. and Leder P.. 1981. Developmental hierarchy of immunoglobulin gene rearrangements in human leukemic pre-B-cells. Proc. Natl. Acad. Sci. USA. 78:7096-7100. Koshland M.E.. 1985. The coming of. age of J chain. Ann. Rev. Immunol. 3:425-453. Krysteva M.A. Mazurier J. and Spik G.. 1976. Ultraviolet difference spectral studies of human serotransferrin and lactotransferrin. Biochim. Biophys. Acta 453:484-493. Krysteva M.A. Mazurier J. Spik G. and Montreuil J.. 1975. Comparative study on histidine modification by diethlpyrocarbonate in human serotransferrin and lactotransferrin. FEBS Lett. 56:337-340. Lamm M.E. Weisz-Carrington P. Roux M.E. McWilliams M. and Phillips-Quagliata J.M.. 1977. Development of the IgA system in the mammary gland. Adv. Exp. Med. Biol. 107:35-42. Landreth K.S. Rosse C. and Clagett J.. 1981. Myelogenous production and maturation of B lymphocytes in the mouse. J. Immunol. 127:2027-2034. Law, B.A. and Reiter, B. 1977. The isolation and bacteriostatic properties of lactoferrin from bovine milk whey. J. Dairy Res. 44:595-599. Leary H.J. and Lecce J.G. 1979. The preferential transport of immunoglobulin G by the small intestine of the neonatal piglet. J. Nutr. 109:458-466. Lecce J.G. Matrone G. and Morgan D.O.. 1961. Porcine neonatal nutrition: Ab-sorption of unaltered non-porcine proteins and polyvinylpyrrolidone from the gut of piglets and the subsequent effect on the maturation of the serum profile. J. Nutr. 73:158-166 Lecce J.G.. 1973. Effect of dietary regimen on cessation of uptake by piglet intestinal epithelium (closure and transport to the blood). J. Nutr. 103:751-756. Lee Y.Z. Sim J.S. Al-Mashikhi S. and Nakai S.. 1988. Separation of immuno-globulins from bovine blood by polyphosphate precipitation and chromatography. J. Agric. Food Chem. 36:922-928. L'Eplattenier F. Murase I. and Martell A.E.. 1967. New multidentate ligands. VI. Chelating tendencies of N,N'-bis(o-hydroxybenzyl)-ethylenediamine diacetic acid. J. Amer. Chem. Soc. 89:837-843. Lewis A.J. Speer V.C. and Haught D.G.. 1978. Relationship between yield and composition of sow's milk and weight gains of nursing piglets. J. Anim. Sci. 47:634-638. 173 Lodinova R. and Jouja U.. 1977. Influence of oral lysozyme administration on serum immunoglobulin and intestinal secretory IgA levels in infants. Acta Paediatr. Scand. 66:709-712. Luk C.K.. 1971. Study of the nature of the metal-binding sites and estimate of the distance between the metal-binding sites in transferrin using trivalent lanthanide ions as fluorescent probes. Biochem. 10:2838-2843. MacGillvray R.T.A. Mendez E. and Brew K.. 1977. Structure and evolution of serum transferrin, in. Proteins of Iron Metabolism, eds. Brown E.B. Aisen P. Fielding J. and Crichton R.R.. pp. 133-141. Grune and Stratton: New York. Maryanski J.L. Pala P. Corradin C. Jordan B.R. and Cerottini J.-C. 1986. H-2 restricted cytolytic T cells specific for HLA can recognize a synthetic HLA peptide. Nature 324:578-582. Masson P.L. and Heremans J.F.. 1971. Lactoferrin in milk from different spe-cies. Comp. Biochem. Physiol. 39B:119-129. Max E.E.. 1984. Immunoglobulins: Molecular genetics, in. Fundamental Immunology, ed. Paul W.E.. pp. 167-204. Raven Press: New York. Mazurier J. Aubert J.-P. Loucheux-Lefevre M.-H. and Spik G.. 1976. Comparative circular dichroism studies of iron-free and iron-saturated forms of human sero-transferrin and lactotransferrin. FEBS Lett. 66:238-242. McCallum I.M. E l l i o t J.I. and Owen B.D. 1977. Survival of colostrum-deprived neonatal piglets fed gamma-globulins. Can. J. Anim. Sci. 57:151-158. McCallum I.M.. 1977. The provision of passive immunity to colostrum-deprived piglets a r t i f i c i a l l y reared in a barn environment. M.Sc. Thesis. University of Saskatchewan. McWilliams M. Phillips-Quagliata J.M. and Lamm M.E.. 1975. Characteristics of mesenteric lymph node cells homing to gut associated lymphoid tissue in syn-geneic mice. J. Immunol. 115:54-58. Mellencamp M.W. McCabe M.A. and Kochan I. 1981. The growth-promoting effect of bacterial iron for serum exposed bacteria. Immunol. 43:483-491. Mendel L. Tavinicek J. and Sterzl J.. 1977. Our experience in rearing germ free piglets: History, contemporary status and outlooks. Acta Vet. Brno. 46: Suppl. 4, 3-11. Merriman C.R. Pulliam L.A. and Kampschmidt R.F.. 1977. Comparison of leucocytic pyrogen and leucocytic endogenous mediator. Proc. Soc. Exp. Biol. Med. 154:224-227. Miedema F. and Melief C.J.M.. 1985. T-cell regulation of human B-cell activation. A reappraisal of the role of interleukin 2. Immunol. Today 6:258-259. 174 Miles A.A. and Khimji P.L. 1975. Enterobacterial chelators of iron: Their occur-rence, detection and relation to pathogenicity. J. Med Microbiol. 8:477-490. Miller E.R. Ullrey D.E. Ackermann I. Schmidt D.A. Luecke R.W. and Hoeffer J.A. 1961. Swine hematology from birth to maturity. II. Erythrocyte population, size and hemoglobin concentration. J. Anim. Sci. 20:890-897. Morisaki I. Michalak S.M. Harmon C.C. Tora M. Hamada S. and McGhee J.R.. 1983. Effective immunity to dental caries: Enhancement of salivary anti-Streptococcus  mutans antibody responses with oral adjuvants. Infect. Immun. 40:577-591. Mostov K.E. Kraehenbuhl J-P. and Blobel. G.. 1979. Receptor-mediated trans-cellular transport of immunoglobulin: Synthesis of secretory component as multiple and larger transmembane forms. Proc. Natl. Acad. Sci. USA 77:7257-7261. Mullock B.M. Jones R.S. and Hinton R.H.. 1980. Movement of endocytic shuttle vesicles from the sinusoidal to the bile canalicular face of hepatocytes does not depend on occupation of receptor sites. FEBS Lett. 113:201-205. N.R.C.. 1978. The nutrient requirements of swine. National Academy of Sciences-National Research Council, Washington, D . C Nagy L.K. Mackenzie T. and Bharucha Z.. 1976a. In vitro studies of the antimicrobial effects of colostrum and milk from vaccinated and unvaccinated pigs. Res Vet. Sci. 21:132-140. Nagy L.K. Bhogal B.S. and Mackenzie T.. 1976b. The effect of colostrum or post coli b a c i l l o s i s on the adhesion of Escherichia c o l i to the small intestine of the pig. Res. Vet. Sci. 21:303-308. Nocek J.E. Braund D.G. and Warner T.G.. 1984. Influence of neonatal colostrum administration, immunoglobulin and continued feeding of colostrum on calf gain, health and serum protein. J. Dairy Sci. 67:319-333. Owen B.D. Bell J.M. and Williams CM.. 1961. Effects of porcine immunoglobulin administration on the survival and serum protein composition of colostrum-deprived piglets reared in a non-isolated environment. Can. J. Anim. Sci. 41:236-252. Owen B.D. and Bell J.M.. 1964. Further studies of survival and serum protein composition in colostrum-deprived pigs reared in a non-isolated environment. Can. J. Anim. Sci. 44:1-7. Parmely M.J. and Beer A.E.. 1977. Colostral cell-mediated immunity and the concept of a common secretory immune system. J. Dairy Sci. 60:655-665. Parmely M.J. Beer A.E. and Billingham R.E.. 1976. In vitro studies on the T-lymphocyte population of human milk J. Exp. Med. 144:358-370. 175 Patt J.A.. 1977. Factors affecting the duration of intestinal permeability of macromolecules in new born animals. Biol. Rev. 52:411-429. Pery P.. 1973. Porcine immunoglobulins. Structure and function. Agricultural research seminar on porcine immunology. Thiveral-Grignon France. Pescowitz M.D. Lunney J.K. and Sachs D.H.. 1985. Murine antiswine T4 and T8 monoclonal antibodies: distribution and effects on proliferative and cytotoxic T c e l l s . J. Immunol. 134:37-44. Pettigrew J.E. Sower A.F. Cornelius S.G. and Moser R.L. 1985. A comparison of isotope dilution and weigh-suckle-weigh methods for estimating milk intake by pigs. Can. J. Anim. Sci. 65:989-992. Phillips-Quagliata J.M. Roux M.E. Arny M. Kelly-Hatfield P. McWilliams M. and Lamm M.E.. 1983. Migration and regulation of B cells in the mucosal immune system. Ann. NY Acad. Sci. 409:194-203. Pierce N.F. and Reynolds H.Y.. 1975. Immunity to experimental cholera. I. Secretory and humoral antitoxin response to local and systemic toxoid admin-istration. J. Infect. Dis. 131:383-389. Pitt J. Barlow B. Heird W.C. and Santulli T.V.. 1974. Macrophages and the protective action of breast milk in necrotizing enterocolitis. Pediat. Res. 8:384. (abs). Plaut A.G.. 1983. The IgAl proteases of pathogenic bacteria. Ann. Rev. Micro-b i o l . 37:603-622. Polis B.D. and Shmukler H.W.. 1953. Crystalline lactoperoxidase. I. Isolation by displacement chromatography. II. Physicochemical and enzymic properties. J. Biol. Chem. 201:475-500. Porter R.R.. 1959. The hydrolysis of rabbit t-globulin and antibodies with crystalline papain. Biochem J. 73:119-126. Porter P. and Chidlow. J.W.. 1979. Response to E. c o l i antigens via local and parenteral routes linking intestinal and mammary immune mechanisms in passive protection against neonatal col i b a c i l l o s i s in the pig. i n . Immunology of breast milk. eds. Ogra P.L. and Dayton D.H.. pp. 73-80. Raven Press: New York. Powanda M.C. and Beisel W.R.. 1982. Hypothesis: Leukocyte endogenous mediator-/endogenous pyrogen/lymphocyte-activating factor modulates the development of nonspecific and specific immunity and affects nutritional status. Amer. J. Clin. Nutr. 35:762-768. Radl J. Schuit H.R.E. Mestecky J. and Hijmans W.. 1973. The origin of monomeric and polymeric forms of IgA in man. Adv. Exp. Med. Biol. 45:57-65. 176 Rainard P. 1986a. Bacteriostasis of Escherichia c o l i by bovine lactoferrin, transferrin and immunoglobulins (IgGi, IgG2, IgM) acting alone or in combin-ation. Vet. Microbiol. 11:103-115. Rainard P. 1986b. Bacteriostatic activity of bovine milk lactoferrin against mastitic bacteria. Vet. Microbiol. 11:387-392. Ravetch J.V. Siebenlist U. Korsmeyer S. Waldmann T. and Leder P.. 1981. Structure of the human immunoglobulin u locus: Characterization of embryonic and rearranged J and D genes Cell 27:583. Regnier J.A. and Kelley K.W.. 1981. Heat- and cold-stress suppresses in vivo and in vitro cellular immune responses of chickens. Amer. J. Vet Res. 42:294-299. Reiter B. and Brock J.H. 1975. Inhibition of Escherichia c o l i by bovine colostrum and post-colostral milk. I. Complement-mediated bactericidal activity of antibodies to a serum susceptible strain of E. c o l i of the serotype 0 111. Immunol. 28:71-82. Reiter B. Brock J.H. and Steel E.D.. 1975. Inhibition of Escherichia c o l i by bovine colostrum and post-colostral milk. II. The bacteriostatic effect of lactoferrin on a serum susceptible and serum resistant strain of E. c o l i . Immunol. 28:83-95. Reiter B. Marshall V.M.E. Bjorck L. and Rosen C.G.. 1976. Nonspecific baterial activity of the lactoperoxidase-thiocyanate-hydrogen peroxide system of milk against Escherichia c o l i and some gram negative pathogens. Infect. Immun. 13:800-807. Reiter B.. 1978. Review of nonspecific antimicrobial factors in colostrum Ann. Rech. Vet. 9:205-224. Reiter B. and Harnulv B.G.. 1982. The preservation of refrigerated and uncooled milk by i t s natural lactoperoxidase system. Dairy Ind. Int. 47:13-19. Reiter B.. 1985a. Interaction between immunoglobulins and innate factors such as lysozyme, lactoferrin, lactoperoxidase. in. Composition and physiological properties of human milk. ed. Schaub J.. pp. 271-281. Elsevier: Amsterdam. Reiter B.. 1985b. The biological significance of the non-immunoglobulins pro-tective proteins in milk: Lysozyme, Lactoferrin, Lactoperoxidase. Dev. Dairy Chem. 3:281-336. Renshaw H.W. Eckblad W.P. Thacker D.L. and Frank F.W.. 1976. Antibacterial host defense: In vitro interaction of bacteria, serum factors and leukocytes from precolostral dairy calves and their dams. Amer. J. Vet. Res. 37:1267-1274. Rice C.E. and Ecuyer C.L.. 1963. Complement titers of naturally and a r t i f i c i a l l y raised piglets. I. In piglets of different birth weights. Can J. Comp. Med. Vet. Sci. 27:157-161. 177 Robinson J.E. Harvey B.A.M. and Soothil J.F.. 1978. Phagocytosis and k i l l i n g of bacteria and yeast by human milk cells after opsonization in aqueous phase of milk. B r i t . Med. J. 1:1443-1445. Rogers H.J.. 1976. Ferric iron and the antibacterial effect of horse 7s antibodies to Escherichia c o l i 0111. Immunol. 30:425-433. Rogers H.J.. 1973. Iron-binding catechols and virulence in Escherichia c o l i . Infect. Immun. 7:445-456. Rosenberg H. and Young I.G.. 1974. Iron transport in the enteric bacteria. In. Microbial Iron Metabolism, ed. Neilands J.B.. pp. 67-82. Academic Press: New York. Rothschild M.F. Zimmerman D.R. Johnson R.K. Venier L. and Warner CM.. 1984. SLA haplotype differences in lines of pigs which differ in ovulation rate. Anim. Blood Groups and Biochem. Genet. 15:155-158. Rothschild M.F. Renard C. Bolet G. Dando P. Vaiman M.. 1986. Effect of swine lymphocyte antigen haplotypes on birth and weaning weights in pigs. Anim. Genet. 17:267-272. Salmon H.. 1987. The intestinal and mammary immune system in pigs. Vet. Immunol. Immunopathol. 17:367-388. Samson R.R. Mirtle C. and McClelland B.L. 1979. Secretory IgA does not enhance the bacteriostatic effects of iron-binding or vitamin B12 binding proteins in human colostrum. Immunol. 38:367-373. Samuelson L.E. Harford J.B. and Klausner R.D.. 1985. Identification of the components of the murine T c e l l antigen receptor complex. Cell 43:223-236. Schlabach M.R. and Bates G.W. 1975. The synergistic binding of anions and Fe 3 + by transferrin. J. Biol. Chem. 250:2182-2188. Scollay R. Butcher E. and Weissman I.L.. 1980. Thymus c e l l migration, quantitative aspects of cellular t r a f f i c from the thymus to the periphery in mice. Eur. J. Immunol. 10:210-218. Scoot A. Owen B.D. and Agar J.L.. 1972. Influence of orally administered porcine immunoglobulins on survival and performance of newborn colostrum-dep-rived pigs. J. Anim. Sci. 35:1201-1205. Sellwood R. Hall G. and Anger H.. 1986. Emigration of polymorphonuclear leucocytes into the intestinal lumen of the neonatal piglet in response to challenge with K88-positive Escherichia c o l i . Res. Vet. Sci. 4:128-135. Senft B. and Klobassa F.. 1971. Untersuchungen uber die kunstliche Aufzucht von Ferkein. Zuchtungskunde 43:371-381. 178 Shen F.W. McDougal J.S. Bard J. and Cort S.P.. 1980. Developmental and communicative interactions of Lyl23 and Lyl cells sets. J. Exp. Med. 151:566-572. Sherman A.R.. 1984. Iron, infection and immunity, in. Nutrition, Disease Re-sistance and Immune Function, ed. Watson R.R.. pp. 251-266. Marcel Dekker: New York. Shevach E.M.. 1984. Macrophages and other accessory c e l l s , i n . Fundamental Immunology, ed. Paul W.E.. pp. 71-107. Raven Press: New York. Shoen I. and Solomon M.. 1962. Control of blood haemoglobin determination by a simple effective method. J. Clin. Pathol. 15:44-46. Silberberg-Sinakin I. Gigli I. Baer R.L. and Thorbecke G.J.. 1980. Langerhans ce l l s : Role in contact hypersensitivity and relationship to lymphoid dendritic cells and to macrophages. Immunol. Rev. 53:203-232. Silverton E.W. Navia M.A. and Davies D.R.. 1977. Three-dimensional structure of an intact human immunoglobulin. Proc. Natl. Acad. Sci. USA 74:5140-5144. Smith L.K. and Schanbacher F.L.. 1977. Lactoferrin as a factor of resistance to infection of the bovine mammary gland. J. Amer. Vet. Med. Assoc. 170:1224-1227. Snodgrass H.R. Kisielow P. Kiefer M. Steinmetz M. and von Boehmer H.. 1985. Ontogeny of the T-cell antigen receptor within the thymus. Nature 314:103-107. Solari R. and Kraehenbuhl J-P.. 1984. Biosynthesis of the IgA antibody receptor: A model for the transepithelial sorting of a membrane glycoprotein. Cell 36:61-67. Solomon J.B.. 1971. Ontogeny of defined immunity in animals, in. Foetal and neonatal immunology, ed. Neuberger A. and Tatum E.L.. pp. 234-306. North Holland: Amsterdam. Spik G. Cheron A. Montreuil J. and Dolby J.M. 1978. Bacteriostasis of a milk-sensitive strain of Escherichia c o l i by immunoglobulins and iron-binding proteins in association. Immunol. 35:663-671. Statistical Analysis System Institute Inc.. 1985. SAS User's Guide: Statistics SAS Institute Inc.: Cary, N.C.. Stephens S. Dolby. J.M. Montreuil J. and Spik G. 1980. Differences in inhibition of the growth of commensal and enteropathogenic strains of Escherichia c o l i by lactotransferrin and secretory immunoglobulin A isolated from human milk. Immunol. 41:597-603. Sterzl J. and Silverstein A.M.. 1967. Developmental aspects of immunity. Adv. Immunol. 6:337-459. 179 S t i f e l F.B. and Vetter R.L. 1967. Effect of a synthetic chelating agent upon forage intake and ruminal fermentation in lambs. J. Anim. Sci. 26:129-135. Stookey L.L. 1970. FerroZine - A new spectrophotometry reagent for iron. Anal. Chem. 42:779-781. Suganuma A. Ishizuka A. Sakiyama Y. Maede Y. and Namioka S.. 1986. B lymphocyte differentiation and suppressor activity by T lymphocytes derived from neonatal and sucking piglets. Res. Vet. Sci. 40:400-405. Tada T.. 1984. Help, suppression, and specific factors, in. Fundamental Immunology, ed. Paul W.E.. pp. 481-517. Raven Press: New York. Talbot R.B. and Swenson M.J.. 1970. Blood volume of pigs from birth through 6 weeks of age. Amer. J. Physiol. 218:1141-1144. Teuwisson B. Masson P.L. Osinski P. and Heremans J.F.. 1972. Metal-combining properties of human lactoferrin. The possible involvement of tyrosyl residues in the binding sites. Spectrophotometric tit r a t i o n . Eur. J. Biochem. 31:239-245. Thomas E.L. Bates K.P. and Jefferson N.M.. 1981. Peroxidase antimicrobial system of human saliva: Requirements for accumulation of hypothiocyanite. J. Dent. Res. 60:785-796. Tizard I.. 1982. Antigen structure and immunogenicity. J. Amer. Vet. Med. Assoc. 181:978-982. Tizard I.. 1986a. Basic immunology-3: The importance of macrophages. Vet. Med. 81:250-257. Tizard I.. 1986b. Basic immunology-2: Reviewing the properties of antibodies. Vet. Med. 81:166-178. Tizard I.. 1986c. Basic immunology-5: The complement cascade. Vet. Med. 81:437-445. Tomasi T.B. and Calvanico M.. 1968. Human secretory T A . Fed Proc. 27:617 (abs.) Trugo N.M.F. Ford J.E. and Sansom B.F.. 1985. Vitamin B12 absorption in the neonatal piglet. I. Studies in vivo on the influence of the vitamin B12-binding protein from sow's milk on the absorption of vitamin B12 and related compounds. Brit. J. Nutr. 54:245-255. Underdown B.J. and Schiff J.M.. 1986. Immunoglobulin A: Strategic defense i n i t i a t i v e at the mucosal surface. Ann. Rev. Immunol. 4:389-417. Vaiman M.. 1987. MHC in farm animals. Anim Genet. 18:7-10 Suppl. 1. Vaiman M. Chardon P. and Cohen D.. 1986. DNA polymorphism in the major histocompatibility complex of man and various farm animals. Anim. Genet. 17:113-133. 180 Van Snick J.L. Masson P.L. and Heremans J.F.. 1974. The involvement of lacto-ferrin in the hyposideremia of acute inflammation. J. Exp. Med. 140:1068-1084. Van Snick J.L. Masson P.L. and Heremans J.F.. 1973. The involvement of b i -carbonate in the binding of iron by transferrin. Biochim. Biophys. Acta 322:231-233. Varley M.A. Maitland A. and Towle A.. 1986. A r t i f i c i a l rearing of piglets: the administration of two sources of immunoglobulins after birth. Anim. Prod. 43:121-126. Voller, A. Bidwell, D.E. and Bartlett, A. 1976. Enzyme immunoassays in diag-nostic medicine. Bull. World Health Organization 53:55-65. von Boehmer H.. 1988. The developmental biology of T lymphocytes. Ann. Rev. Immunol. 6:309-326. von Boehmer H.. 1986. The selection of the a, fi heterodimeric T-cell receptor for antigen. Immunol. Today 7:333-336. Waldron J.A. Horn R.G. and Rosenthal A.S.. 1973. Antigen-induced proliferation of guinea pig lymphocytes in vitro: Obligatory role of macrophages in the recognition of antigen by immune T-lymphocytes J. Immunol. 111:59-64. Walker W.A. and Isselbacher K.J.. 1974. Uptake and transport of macromolecules by the intestine: possible role in c l i n i c a l disorders. Gastroenterology 67:531-550. Warner R.C. and Weber I.. 1953. The metal-combining properties of conalbumin J. Amer. Chem. Soc. 75:5094-5101. Wasserfall F. and Teuber M.. 1979. Action of eggwhite lysozyme on Clostridium  tyrobutricum. Appl. Environ. Microbiol. 38:197-199. Waterhouse A. and Mullen W.M.A.. 1980. Reinclusion of an active lactoperoxidase system in a milk substitute diet for calves. Anim. Prod. 30:458 (abs.) The Webster New American Dictionary. 1972. eds. Morehead A. and Morehead L.. Signet:Chicago. Weicker J. and Underdown B.J.. 1975. A study of the association of human se-cretory component with IgA and IgM proteins. J. Immunol. 114:1337-1344. Weinberg E.D.. 1974. Iron and susceptibility to infectious disease. Science 184:952-956. Weinberg E.D.. 1984. Iron witholding: A defense against infection and neoplasia. Physiol. Rev. 64:65-102. Weiner R.E. and Szuchet S.. 1975. The molecular weight of bovine lactoferrin. Biochim. Biophys. Acta 393:143-147. 181 Wilson M.R.. 1972. The influence of preparturient intramammary vaccination on bovine mammary secretions. Antibody activity and protective value against Escherichia c o l i enteric infections. Immunol. 23:947-955. Woodworth R.C. Marallee K.G. and Williams R.J.P.. 1970. Perturbations of the proton magnetic resonance spectra of conalbumin and siderophilin as a result of binding Ga3 + or Fe 3 +. Biochem. 9:839-842. Yague J. White J. Coleclough C. Kappler J. Palmer E. and Marrack P.. 1985. The T c e l l receptor: The a and fi chains define idiotype and antigen and MHC specificity. Cell 42:81-87. Yamashita U. and Schevach E.M.. 1977. The expression of la antigens on immunocompetent cells in the guinea pig. J. Immunol. 119:1584-1588. Young G.A. Underdahl N.R. and Hinz R.W.. 1955. Procurement of baby pigs by hysterectomy. Amer. J. Vet Res. 16:123-131. Young G.A. and Underdahl N.R.. 1953. Isolation units for growing baby pigs without colostrum. Amer. J. Vet. Res. 14:571-574. Zimmerman D.R. Speer V.C. Hays B.W. and Catron D.V.. 1959. Injectable iron-dextran and several oral iron treatments for the prevention of iron-deficiency anemia of baby pigs. J. Anim. Sci. 18:1409-1415. 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098280/manifest

Comment

Related Items