key: cord-022544-7jn4ns6x
authors: Lawrence, Robert M.
title: Host-Resistance Factors and Immunologic Significance of Human Milk
date: 2010-12-27
journal: Breastfeeding
DOI: 10.1016/b978-1-4377-0788-5.10005-7
sha: 
doc_id: 22544
cord_uid: 7jn4ns6x

nan

Some of the most dramatic and far-reaching advances in the understanding of the immunologic benefits of human milk have been made using newer techniques to demonstrate the specific contribution of the numerous "bioactive factors" contained in human milk (Table 5 -1). The multifunctional capabilities of the individual factors, the interactive coordinated functioning of these factors, and the longitudinal changes in the relative concentrations of them for the duration of lactation make human milk unique. The immunologically active components of breast milk make up an important aspect of the host defenses of the mammary gland in the mother; at the same time, they complement, supplement, and stimulate the ongoing development of the infant' s immune system. [107] [108] [109] The explosion of research on all the immunologic properties and actions of breast milk in the last 10 years makes it impossible to summarize all the important aspects of what we now know about the immunologic benefits of breast milk. The recently developed technologies of genomic studies using microarrays and proteomics promise to continue this rapid expansion of knowledge on the biology of the mammary gland, human milk, and the infant's developing immune system. This chapter emphasizes the important concepts of these immunologic benefits and refers the interested reader to the most recent literature for more extensive information on the many specific components.

The immunologic benefits of human milk can be analyzed from a variety of perspectives:

1. Reviewing the published information on the protection of infants from specific infections that compares breastfed and formula-fed infants.

developing immune systems and the actions of bioactive factors provided in breast milk. 3. Examining the proposed function of the active components contained in human milk: antimicrobial, antiinflammatory, and immunomodulating. 4. Considering the nature of the different factors: soluble, cellular, and hormone-like. 5. Examining the contribution of breast milk to immune function of mammary glands and infants as an evolutionary process. 6. Determining the site of the postulated action of the specific factors (e.g., in the breast or in the infant) at the mucosal level (respiratory tract or gastrointestinal [GI] tract) or at the systemic level. 7. Classifying the factors relative to their contribution to the constitutive defenses (innate immunity) versus the inducible defenses (adaptive immunity) of the infant' s immune system. 8. Clarifying the mechanism of action of the proposed immunologic benefit (the mucosalassociated lymphoid tissue formation of the CHAPTER 5

bioactive factors at the level of the mucosa and their subsequent action at the breast or in an infant). 9. Considering the contribution of human milk to the development of an infant' s immune system relative to potential long-term immunologic benefits, such as protection against allergy, asthma, autoimmune disease, or inflammatory bowel disease.

The protective effect of breast milk against infection was documented as early as 1892 in the medical literature by data proving that milk from various species, including humans, was protective for offspring, containing antibodies against a vast number of antigens. 271 Veterinarians have long known the urgency of offspring receiving the early milk of the mother. Death rates among human newborns not suckled at the breast in the Third World are at least five times higher than among those who receive colostrum and mother' s milk. The evidence that a lack of breastfeeding and poor environmental sanitation have a pernicious synergistic effect on infant mortality rate has been presented by Habicht 101 after studying 1262 women in Malaysia.

The evidence that breastfeeding protects against infections in the digestive and respiratory tracts has been reported for several decades. 269 However, many of the older studies were criticized for flawed methodology and because they were performed in "developing countries," where the risk for infection due to poor sanitation was expected to be higher. 13, 101, 116 Various researchers have proposed specific criteria for assessing the methodology of studies reporting on the protective effects of breast milk, clearly identifying measurable outcomes and the definition of breastfeeding, with other methods to limit bias and to control for confounding variables. 13, 46, 147, 149 More recent studies, which have incorporated many of the proposed methodologic criteria, continue to document that breastfeeding protects infants against diarrhea, respiratory infections, and otitis media.* Individual papers report protection against urinary tract infections and neonatal sepsis. 6, 109, 220, 275 Several papers document the decreased risk for dying in infancy associated * References 7, 15, 43, 51, 55, 122, 167, 199, 208, 223, 226, 243, 278. with exclusive or predominant breastfeeding in Pakistan, Peru, Ghana, India, Nepal, and Bangladesh. 4, 6, 9, 62, 179 A systematic review by the Bellagio Child Survival Study Group predicted that exclusive breastfeeding for 90% of all infants through 6 months of age could prevent 13% of the childhood deaths occurring younger than 5 years of age. 133 A recent review on human breast milk documents the evidence for protection against infectious diseases from breastfeeding for resource-rich and resourcepoor countries. 153 

One of the important considerations relative to measuring the immunologic benefits of breast milk is the exclusivity and duration of breastfeeding. The basic concept is identifying a dose-response relationship between the amount of breast milk received by an infant during the period of observation and the immunologic benefit gained with greater exclusivity and duration equaling greater volume of breast milk or "dose." Dr. Labbok and Krasovec 149 have carefully defined breastfeeding in terms of the patterns of breastfeeding relative to the amount of supplementation with formula or other fluids or foods (full/nearly full, medium or equal, low partial, or token) to standardize the use of equatable terms in different studies. Box 5-1 outlines these definitions of the "amount" of breastfeeding. 153 Raisler et al 227 referred to a doseresponse relationship when they studied the effect of "dose" of breast milk on preventing illness in more than 7000 infants. "Full breastfeeding" was associated with the lowest rates of illness (diarrhea, cough, or wheeze) and even children with "most" or "equal" breastfeeding had evidence of lower odds ratios of ear infections and certain other illnesses. A number of other long-term studies demonstrated greater protection from infection with increased exclusivity of breastfeeding and durations of at least 3 months.* A couple papers demonstrated a "dose" effect relative to decreased occurrence of late onset sepsis in very low-birth-weight infants 73 and premature infants 245 associated with the infants receiving at least 50 mL/kg per day of mother' s milk compared with receiving other nutrition. The current recommendations from the American Academy of Pediatrics reinforce the importance of the doseresponse relationship between breastfeeding and the benefits of breastfeeding when it recommends exclusive breastfeeding for the first 6 months of life and at least partial breastfeeding after the introduction of solid foods for an additional 12 months or longer. 1 Another important consideration relative to exclusive breastfeeding is the potential effect of other foods and fluids in an infant' s diet that could negatively influence immunologic benefits and infection-protective effects at the level of the GI mucosa.

The human immune system begins forming and developing in the fetus. Newborn infants' immune systems are immature and inadequate at birth. Immune systems rapidly adapt in the postnatal period related to the natural maturation of the skin and mucosal barriers and in response to the exposure of infants to inhaled and ingested antigens and microbial agents in the extrauterine environment. Infants' immune systems develop throughout at least the first 2 years of life. Overall, infants have limited abilities to respond effectively and quickly to infectious challenges, which explains infants' ongoing susceptibility to infections. † Box 5-2 lists most of the better understood deficiencies in infants' immune systems. An extensive discussion of these developmental immune deficiencies affecting infants is presented by Lawrence and Pane. 153 The B-lymphocytes and immunoglobulin production are deficient in the amount and specificity of antibodies produced. There is limited isotype switching and slow maturation of the antibody response to specific antigens (polysaccharides). 114, 177 The systemic cell-mediated immune response, including complement components, and both the "classical" and alternative pathways have limitations for complement activation. 2, 63, 251, 277 Numerous immune components are produced in limited amounts in infancy, including complement, interferon-γ, secretory immunoglobulin (Ig) A, interleukins (IL-3, IL-6, IL-10), tumor necrosis factor (TNF)-α, lactoferrin, and lysozyme. 45, 87 Relative to these various immune deficits in infants, one can find various bioactive and immunomodulating factors in breast milk that are potentially capable of complementing and enhancing the development of infants' mucosal and systemic immune systems. 87, 110 This concept of bioactive and immunomodulating factors in breast milk is an important area of evolving research that has been extensively reviewed in the literature. 87, 88, 110, 144 The most intense focus of this research centers on the effects of human milk on infants' GI tract. 85, 201 

The bioactive factors being studied are as diverse as proteins (lactoferrin, lysozyme, etc.), hormones (erythropoietin, prolactin, insulin, etc.), growth factors (epithelial growth factor, insulin-like growth factor, etc.), neuropeptides (neurotensin, somatostatin, etc.), cytokines (TNF-α, IL-6, etc.), antiinflammatory agents (enzymes, antioxidants, etc.), and nucleotides (see Table 5 -1). In the past, it was adequate to point to the lists of factors (especially immunoglobulins) to "explain" the immunologic benefit of breast milk. Today, it is necessary to understand the multifunctional and dynamic action of these individual factors, their specific mechanisms of action on the innate, the adaptive, and the mucosal immune system, and their role in direct infection protection, in the normal development of infants' immune systems, and their contribution against potentially harmful inflammation.

From an evolutionary perspective, maternal antibody is transmitted to the fetus by different pathways in different species. 87, 159, 261 An association has been recognized between the number of placental membranes and the relative importance of the placenta and the colostrum as sources of antibodies. By this analysis, horses, with six placental membranes, pass little or no antibodies transplacentally and rely totally on colostrum for protection of foals. Humans and monkeys, having three placental membranes, receive more of the antibodies via the placenta and less from the colostrum. The transfer of IgG in humans is accomplished by active transport mechanism of the immunoglobulin across the placenta. Secretory IgA (sIgA) immunoglobulins are found in human milk and provide local protection to the mucous membranes of the GI tract. Other investigations have established that the mammary glands and their secretion of milk are important in protecting the infant not only through the colostrum but also through mature milk from birth through the early months of life.

Although the predominance of IgA in human colostrum and milk had long been described, the importance of this phenomenon was not fully appreciated until the discovery that IgA is a predominant immunoglobulin present in mucosal secretions of other glands in addition to the breast.

Mucosal immunity has become the subject of extensive research. 21, 22, 120, 201 It is clear that considerable traffic of cells occurs between mucosal epithelia and secretory or lymphoid tissue sites. The data support the concept of a general system of mucosa-associated lymphoid tissue (MALT), which includes the gut, lung, mammary gland, salivary and lacrimal glands, and the genital tract ( Figure  5 -3). Through the immune response of MALT, a reaction to an immunogen at a mucosal site may be an effective means of producing immunity at distant sites. Antibodies against specific antigens found in milk have also been found in the saliva, evidence for transfer of protection to two different distant sites simultaneously. Evidence suggests that the mammary glands may act as extensions of the gut-associated lymphoid tissue (GALT) and possibly the bronchiole-associated lymphoid tissue. The ability of epithelial surfaces exposed to the external environment to defend against infectious agents has been well documented for the GI, genitourinary, and respiratory tracts. 142 sIgA and secretory IgM (sIgM) produced through the adaptive response of the mucosal-lymphoid immune system act by blocking colonization with pathogens and limiting the passage of harmful antigens across the mucosal barrier. Activated B cells and cytokines pass to the mammary gland where they contribute to the production of sIgA in breast milk. Direct contact between the antigen and the lymphoid cells of the breast is unlikely. 202 Peyer patches, tonsils, and other MALT structures appear to be well developed at birth. 24 Nevertheless, the actual effective production of sIgA to various antigens presented to infants' mucosal surfaces (respiratory and GI tracts) is still inadequate to protect against infection. A breastfeeding infant, as part of the maternal-infant dyad exposed to the same antigens via their mucosal services, can receive protective sIgA and sIgM in the mother' s breast milk produced by the mother' s MALT ( Figure 5-1) .

The protective properties of human milk can be divided into cellular factors and humoral factors for facility of discussion, although they are closely related in vivo. A wide variety of soluble and cellular components and hormone-like agents have been identified in human milk and colostrum (see Table  5 -1). Although the following discussion separates these elements, it is important to emphasize that the constituents of human milk are multifunctional and their functioning in vivo is interactive and probably coordinated and complementary.

Cells are an important postpartum component of maternal immunologic endowment. More than 100 years ago, cell bodies were described in the colostrum of animals. As with much lactation research, further study of colostral corpuscles was undertaken by the dairy industry for commercial reasons in the early 1900s. This research afforded an opportunity to make major progress in the understanding of cells in milk. Initially, it was thought that these cells represented a reaction to infection in the mammary gland and were even described as "pus cells."

It has become clear that the cells of milk are normal constituents of colostrums in all species. Cells include macrophages, lymphocytes, neutrophils, and epithelial cells, and they total approximately 4000/mm 3 . Cell fragments and epithelial cells were examined by electron microscope in fresh samples from 30 women by Brooker. 27 He found that the membrane-bound cytoplasmic fragments in the sedimentation pellet outnumbered intact cells. The fragments were mostly from secretory cells that contained numerous cisternae of rough endoplasmic reticulum, lipid droplets, and Golgi vesicles containing casein micelles. Secretory epithelial cells were found in all samples and, after the second month postpartum, began to outnumber macrophages. Ductal epithelial cells were about 1% of the population of cells for the first week or so and then disappeared. All samples contained squamous epithelial cells originating from galactophores and the skin of the nipple.

Living leukocytes are normally present in human milk. 142 The overall concentration of these leukocytes is of the same order of magnitude as that seen in peripheral blood, although the predominant cell in milk is the macrophage rather than the neutrophil. Macrophages compose about 90% of the leukocytes, and 2000 to 3000/mm 3 are present. Lymphocytes make up about 5% to 10% of the cells (200 to 300/mm 3 ), which is a much lower concentration than in human blood. 91 The number of cells found in human milk increases with mastitis. Both large and small lymphocytes are present. By indirect immunofluorescence with anti-T-cell antibody to identify thymus-derived lymphocytes, it has been shown that 50% of human colostral lymphocytes are T cells, and in human milk up to 80% of the lymphocytes are T cells. 276 Immunofluorescence procedures to detect surface immunoglobulins characteristic of B lymphocytes identified 34% as B lymphocytes.

The number of leukocytes and the degree of mitogenic stimulation of lymphocytes sharply decline during the first 2 or 3 months of lactation to essentially undetectable levels, according to Ogra and Ogra ( Figure 5 -2). 92 Enumeration of the total cell numbers in milk has been difficult, but when various techniques are compared (Coulter electronic particle counter, visual cell counting with special stains, filter trapping with fluorescent detection, and automated fluorescent cell counting), stains for deoxyribonucleic acid (DNA) were superior to the other techniques.

Macrophages are large-complex phagocytes that contain lysosomes, mitochondria, pinosomes, ribosomes, and a Golgi apparatus. The monocytic phagocytes are lipid laden and were previously called the colostral bodies of Donne. They have the same functional and morphologic features as phagocytes from other human tissue sources. These features include ameboid movement, phagocytosis of microorganisms (fungi and bacteria), killing of bacteria, and production of complement components C3 and C4, lysosome, and lactoferrin. Other milk macrophage activities include the following 219 :

Phagocytosis of latex, adherence to glass Secretion of lysozyme, complement components C3b-mediated erythrocyte adherence IgG-mediated erythrocyte adherence and phagocytosis Bacterial killing Inhibition of lymphocyte mitogenic response Release of intracellular IgA in tissue culture Giant cell formation Interaction with lymphocytes Data suggest these macrophages also amplify T-cell reactivity by direct cellular cooperation or by antigen processing. The colostral macrophage has been suggested as a potential vehicle for the storage and transport of immunoglobin. A significant increase in IgA and IgG synthesis by colostral lymphocytes, when incubated with supernatants of cultured macrophages, has been reported. 221 The macrophage may also participate in the biosynthesis and excretion of lactoperoxidase and cellular growth factors that enhance growth of intestinal epithelium and maturation of intestinal brush-border enzymes.

The mobility of macrophages is inhibited by the lymphokine migration inhibitor factor, which is produced by antigen-stimulated sensitized lymphocytes. The activities of macrophages have been demonstrated in both fresh colostrum and colostral cell culture, and certain functions are altered compared with their counterpart in human peripheral blood.

The highest concentration of cells occurs in the first few days of lactation and reaches more than a million per milliliter of milk.

Colostrum (1 to 4 days postpartum) contains 10 5 to 5 × 10 6 leukocytes/mL, and 40% to 60% are polymorphonuclear cells (PMNs). Mature milk (i.e., after 4 days) has fewer cells (see Figure 5 -3), approximately 10 5 /mL with 20% to 30% PMNs. After 6 weeks, few PMNs are present. The functions of the PMNs normally include microbial killing, phagocytosis, chemotactic responsiveness, stimulated hexose monophosphate shunt activity, stimulated nitroblue tetrazolium dye reduction, and stimulated oxygen consumption. 33 When milk PMNs are compared with those in the serum, their activity is often less than that of serum PMN cells. Whether milk PMNs actually perform a role in protection of the infant has been studied by many investigators using many techniques. Briefly, animal studies have shown that (1) the mammary gland is susceptible to infection in early lactation, (2) a dramatic increase in PMNs occurs with mammary inflammation, and (3) in the presence of peripheral neutropenia during chronic mastitis, severe infection of the gland occurs. This implies, according to Buescher and Pickering, 33 that the primary function of milk PMNs is to defend the mammary tissue per se and not to impart immunocompetence to the newborn. This may explain the presence of large numbers of PMNs that are relatively hypofunctional early and then disappear over time.

Evidence shows that neutrophils found in human milk demonstrate signs of activation, including increased expression of CD11b (an adherence glycoprotein), decreased expression of L-selectin, spontaneous production of granulocyte-macrophage colony-stimulating factor (GM-CSF), and the ability to transform into CD1 + dendritic cells (DCs). 125 Human milk macrophages have the morphology and motility of activated cells. The movement of these cells in a three-dimensional system is greater than that of monocytes, their counterparts in peripheral blood. Such activated neutrophils may play a role in phagocytosis at the level of the mucosa of the GI tract, supplementing infants' poor ability to recruit phagocytes to that site. 137

Both T and B lymphocytes are present in human milk and colostrum and are part of the immunologic system in human milk. T cells are 80% of the lymphocytes in breast milk. Human milk lymphocytes respond to mitogens by proliferation, with increased macrophage-lymphocyte interaction and the release of soluble mediators, including migration inhibitor factor. Cells destined to become lymphopoietic cells are derived from two separate influences, the thymus (T) and the bursa (B) or bursal equivalent tissues. The population of the B cells makes up the smaller part of the total. They synthesize IgA antibody. The term B cell is derived from its origination in a different anatomic site from the thymus; in birds, it has been identified as the bursa of Fabricius. The B cells can be identified by the presence of surface immunoglobulin markers. The B cells in human milk include cells with IgA, IgM, and IgG surface immunoglobulins. B cells transform into plasma cells and remain sessile in the tissues of the mammary gland.

More rapid mitotic activity occurs in the thymus gland than in any other lymphatic organ, yet 70% of the cells die within the cell substance. The thymus is the location for much of the T-cell differentiation and selection and plays a major role in the development of infants' immune systems. Thymosin has been identified as a hormone produced by thymic epithelial cells to expand the peripheral lymphocyte population. After emergence from the thymus gland, T cells acquire new surface antigen markers. The T cells circulate through the lymphatic and vascular systems as long-lived lymphocytes, which are called the "recirculating pool." They then populate restricted regions of lymph nodes, forming thymic-dependent areas. 276 It is interesting to note that exclusively breastfed infants have a significantly larger thymus than formula-fed infants at 4 and 10 months. 111 The significance of the lymphocytes in human milk in affording immunologic benefits to breastfed infants continues to be investigated. It is suggested that lymphocytes can sensitize, induce immunologic tolerance, or incite graft-versus-host reactions. According to Head and Beer, 115 lymphocytes may be incorporated into sucklings' tissues, achieving short-term adoptive immunization of the neonate.

Studies of the activities of lymphocytes have been carried out by a number of investigators who collected samples of milk from lactating women at various times postpartum, examined the number of cell types present, and then studied the activities of these cells in vitro. 138, 142 Ogra and Ogra 203 collected samples from 200 women and measured the cell content from 1 through 180 days (see Figure  5 -3). They then compared the response of T lymphocytes in colostrum and milk with that of the T cells in the peripheral blood. T-cell subpopulations have also been shown by surface epitopes to be similar to those in the peripheral blood.

The greatest number of cells appeared on the first day, with the counts ranging from 10,000 to 100,000/mm 3 for total cells. By the fifth day, the count had dropped to 20% of the first day' s count. In addition, the number of erythrocyte rosette-forming cells was determined by using sheep erythrocyterosetting technique. The erythrocyte rosette formation lymphocytes constituted a mean 100/mm 3 on the first day and 1 ⁄10 of that by the fifth day.

At 180 days, total cells were 100,000/mm 3 , lymphocytes were 10,000/mm 3 , and erythrocyte rosette formation lymphocytes were 2000/mm 3 . The investigators compared the values with those in the peripheral blood of each mother; the levels remained essentially constant. 202 In a similar study, Bhaskaram and Reddy 18 sampled milk over time from 74 women and found comparable cell concentrations. They examined the bactericidal activity of the milk leukocytes and found it to be comparable with that of the circulating leukocytes in the blood, irrespective of the stage of lactation or state of nutrition of the mother.

Ogra and Ogra 203 also studied the lymphocyte proliferation responses of colostrum and milk to antigens. Their data show response to stimulation from the viral antigens of rubella, CMV, and mumps. Analysis of cell-mediated immunity to microbial antigens shows milk lymphocytes are limited in their potential for recognizing or responding to certain infectious agents compared with cells from the peripheral circulation. This is thought to be an intercellular action and not caused by lack of external factors. In contrast, the T cells and B cells have been shown to have unique reactivities not seen in peripheral blood.

Colostral lymphocytes are derived from mature rather than immature T-cell subsets. The distribution of T-cell subsets in colostrum includes both CD4 + and CD8 + cells. 231 The distribution of CD4 cells in colostrum and human milk is lower than in the serum, and fewer CD4 cells exist than CD8 cells. 276 The percentage of CD4 cells is higher than in the serum of either postpartum donors or normal control subjects. No correlation exists with length of gestation and number of cells (normal blood usually contains twice as many CD4 + as CD8 + lymphocytes). 134 Parmely et al 212 partially purified and propagated milk lymphocytes in vitro to study their immunologic function. Milk lymphocytes responded in a unique manner to stimuli known to activate T lymphocytes from the serum. The authors found milk lymphocytes to be hyporesponsive to nonspecific mitogens and histocompatibility antigens on allogenic cells in their laboratory. They found them unresponsive to Candida albicans. Significant proliferation of lymphocytes occurred in response to K 1 capsular antigen of Escherichia coli. 118 Lymphocytes from blood failed to respond to the same antigen. This supports the concept of local mammary tissue immunity at the T-lymphocyte level.

More recent experiments in rodents have provided evidence that T lymphocytes that are reactive to transplantation alloantigens can adoptively immunize a suckling newborn. Foster nursing experiments performed in rodents have shown that newborn rats exposed to allogenic milk manifested alterations in their reactivity to skin allografts of the foster mother' s strain. In animals, mothers may give their suckling newborn immunoreactive lymphocytes. The influence of maternal milk cells on the development of neonatal immunocompetence has been demonstrated in several different immunologic contexts. Congenitally, athymic nude mice nursed by their phenotypically normal mothers or normal foster mothers had increased survival. The mothers contributed their T-cell-helper activity to the suckling newborn.

Colostral lymphocytes proliferate in response to various mitogens, alloantigens, and conventional antigens. Colostral cells survive in the neonatal stomach and in the gut of experimental animals, some remaining viable in the upper GI tract for a week. No evidence, however, indicates that transepithelial migration takes place when neonatal mice are foster-nursed by newly delivered animals whose colostral cells were tagged with 3 H-thymidine. 33 Cells in human milk have been studied using the same markers employed with cells in the peripheral blood; 80% of the lymphocytes are T cells that are equally distributed between CD4 + and CD8 + subpopulations, and their T-cell receptors are principally of the α/β type. CD4 + cells are common leukocyte cells of the helper and suppressorinducer subsets, and CD8 + cells are leukocytes of the cytotoxic and noncytotoxic subsets. T cells in human milk are presumed activated because they display increased phenotypic markers of activation including HLA-DR and CD25 (IL-2 receptor). The majority of T cells in human milk are CD45RO + , consistent with effector and memory T cells. 236, 276 These cells are effective producers of interferon-γ, which is consistent with their phenotypic features.

Here again, human milk may supplement the infant with a functioning immune cell to compensate for an identified deficiency in the infant, a paucity of memory T cells.

Juto 134 studied the effect of human milk on B-cell function. Cell-free, defatted, filtered colostrum as well as mature breast milk showed an enhancing effect on B-cell proliferation and generation of antibody secretion. This was not seen with formula. Juto suggested that this could represent an important immunologic mechanism. Goldblum et al 82 were able to show a B-cell response in human colostrum to E. coli given to the mother orally, which was not accompanied by a systemic response in the mother. This suggests that the breast and breast milk reflect sites of local humoral or cell-mediated immunity, which were initially induced at a distant site such as the gut and transferred via reactive lymphoid cells migrating to the breast. Head and Beer 115 provided a scheme to describe this mechanism (see Figure 5 -1). The diagram depicts the progeny of specifically sensitized lymphocytes that originated in GALT, specifically Peyer patches, as they migrate to the mammary gland. As they infiltrate the mammary gland and its secretion, they supply the breast with immune cells capable of selected immune responses. Ogra and Ogra 202, 203 suggest that the cells may selectively accumulate in the breast during pregnancy. The responses of milk cells and their antibodies are not representative of an individual' s total immunity. 212 Most of these immunocompetent cells, initially stimulated in GALT, recirculate to the external mucosal surface and populate the lamina propria as antibody-producing plasma cells. A substantial number of these antigen-sensitized cells selectively home-in to the stroma of the mammary glands and initiate local IgA antibody synthesis against the antigens initially encountered in the respiratory or intestinal mucosa. 18 More recent work on human-milk-derived B cells demonstrates that breast milk contains activated memory B cells, different than those in the blood. These cells express mucosal adhesion molecules (α 4 β 7 -/+ , α4β 1 + , CD44 + , CD62L -) suggesting origin in the mammary gland, but similar to GALT-associated B cells. 265 The mucosae-associated epithelial cytokine CCL28 may contribute to migration of and retention of these cells in the mammary gland. 274 This information supports the concept of the mammary gland as an effector site of the mucosal immune system.

The accumulated epidemiologic research support the concept that colostrum and milk provide human infants with immunologic benefits. Both T and B lymphocytes found in breast milk are reactive against organisms invading the intestinal tract. However, the proof of specific viral or bacterial protection secondary to the action of immunologically active B cells has not been demonstrated.

Although it is clear that cells are provided in the colostrum and milk, the effectiveness and impact of these cells on the neonate depend on their ability to survive in the GI tract. It has been demonstrated in several species, including humans, that the pH of the stomach can be as low as 0.5, but the output of hydrochloric acid is minimal for the first few months, as is the peptic activity. Immediately after a feeding begins, the pH rises to 6.0 and returns to normal in 3 hours. The cells from milk tolerate this. Studies in rats have also shown that intact nucleated lymphoid cells are found in the stomach and intestines. 16 These cells, when removed from rat stomachs, are capable of phagocytosis. Lymphoid cells in milk have been shown to traverse the mucosal wall.

When human milk is stored, however, the cellular components do not tolerate heating to 63° C (145.4° F), cooling to −23° C (−9.4° F), or lyophilization. Although a few cells may be identified in processed milk, they are not viable. 83 

Cregan et al 49 have reported the presence of mammary stem cells in human breast milk based on the demonstration of the cytokeratin 5 mammary stem cell marker. Additionally cells from breast milk were analyzed after culturing which showed both a multipotent stem cell marker, nestin, and the cytokeratin 5 marker. Although human milk may serve as a source of mammary stem cells in the future, no evidence of an immunologic role for these cells in developing infants currently exists.

All classes of immunoglobulins are found in human milk. The study of immunoglobulins has been enhanced through the techniques of electrophoresis, chromatographics, and radioimmunoassay. More than 30 components have been identified; of these, 18 are associated with proteins in the maternal serum, and the others are found exclusively in milk. The concentrations are highest in the colostrum of all species, and the concentrations change as lactation proceeds. 173 IgA, principally sIgA, is highest in colostrum. Although postpartum levels fall throughout the next 4 weeks, substantial levels are maintained throughout the first year, during gradual weaning between 6 and 9 months, and even during partial breastfeeding (when infant receives solid foods) in the second year of life (Figures 5-4 and 5-5 and Table 5 -2). Specific sIgA antibodies to E. coli persist through lactation and may even increase (see Figure 5 -4).

The main immunoglobulin in human serum is IgG; IgA content is only one fifth the level of IgG. In milk, however, the reverse is true. IgA is the most important immunoglobulin in milk, not only in concentration but also in biologic activity. sIgA is likely synthesized in the mammary alveolar cells 256 or by lymphocytes that have migrated from Peyer patches in the GI tract or from lymphoid tissue in the respiratory tract via the lymphatics to the breast. Cytokines cause isotype switching of local IgM + B cells to become IgA + B lymphocytes. 85, 248, 272 These isotype switched cells travel to the breast where they are transformed into plasma cells producing secretory, dimeric IgA. It is through this "enteromammary" pathway that the mother provides increased amounts of sIgA to the infant against the microorganisms present in the mother' s and infant' s environment. 261 Brandtzaeg et al 22 have proposed a model for the transport of IgA (polymeric) and IgM (pentameric), produced by plasma cells, across the secretory epithelium with the formation of sIgA and IgM through binding with the secretory component attached to the epithelial membrane. This occurs in the membrane of mammary epithelial cells during lactation. 23, 93 Quantitative determinations of immunoglobulins in human milk were made from milk collected at birth to as long as 27 months postpartum by Peitersen et al 213 IgA is the predominant immunoglobulin in breast milk, constituting 90% of all the immunoglobulins in colostrum and milk.

Ogra and Ogra [202] [203] [204] studied the serum of postpartum lactating mothers and nonpregnant matched control subjects and noted that the individual and mean concentrations of all Ig classes were lower in the postpartum subjects. The levels were statistically significant for IgG; they were 50 to 70 mg higher in the nonpregnant women.

Immunoglobulin levels, particularly IgA and IgM, are very high in colostrum and drop precipitously in the first 4 to 6 days, but IgG does not show this decline. The volume of mammary secretion, however, increases dramatically in this same period; thus the absolute amounts of immunoglobulins remain more nearly constant than it would first appear. Local production and concentration of IgA and probably IgM may take place in the mammary gland at delivery.

IgE and IgD have also been measured in colostrum and milk. Using radioimmunoassay techniques, colostrum was found to contain concentrations of 0.5 to 0.6 IU/mL IgE in 41% of samples and less in the remainder. 10 IgD was found in all samples in concentrations of 2 to 2000 mg/dL. Plasma levels were poorly correlated. The findings suggest possible local mammary production rather than positive transfer. The question of whether IgE or IgD antibodies in breast milk have similar specificities for antigens as the IgA antibodies in milk remains unanswered. 171 Keller et al 139 examined the question of local mammary IgD production and its possible participation in a mucosal immune system by comparing colostrum and plasma levels of total IgD with specific IgD antibodies. From their work comparing colostrum/plasma ratios for IgG, IgD, and albumin and measuring IgD against specific antigens, the authors reported evidence for IgD participation in the response of the mucosal immune system, with increases in total IgD and IgD against specific antigens found in colostrum.

To address the question of total quantities of immunologic components secreted into human milk per day and available to an infant, Butte et al 36 Lysozyme, in contrast, rose during the same period in total amount available and amount per kilogram per day. The authors 36 suggest that production and secretion of these immunologic factors by the mammary gland may be linked to the catabolism of the components at an infant' s mucosal tissues. When the concentrations of sIgA, IgG, IgM, α 1antitrypsin, lactoferrin, lysozyme, and globulins C3 and C4 were compared in relationship to parity and age of the mother, no consistent trend was observed. When maturity of the pregnancy was considered, however, mean concentrations of all these proteins were higher, except for IgA, when the delivery was premature. Because several proteins in human milk have physiologic function in infants, Davidson and Lönnerdal 53 examined the survival of human milk proteins through the GI tract. Crossed immunoelectrophoresis showed that three human milk proteins transversed the entire intestine and were present in the feces: lactoferrin, sIgA, and α 1 -antitrypsin. Miranda et al 176 reported on the effect of maternal nutritional status on immunologic substances in human colostrum and milk. Maternal malnutrition was characterized as lower weight-to-height ratio, creatine/height index, total serum proteins, and IgG and IgA. In malnourished mothers, the colostrum contained one third the normal concentration of IgG, less than half the normal level of albumin, and lower IgA and complement C4. Lysozyme, complement C3, and IgM levels were normal. Levels improved with development of mature milk and improvement in maternal nutrition. According to one report in 2003, moderate exercise during lactation does not affect the levels of IgA, lactoferrin, or lysozyme in breast milk. 165 Immunologic components contained in human milk during the second year of lactation become a significant point as more infants are nursed longer. For a longitudinal study of lactation into the second year by Goldman et al, 89 women were included who had fully breastfed their infants for 6 months to a year, and were continuing to partially breastfeed. Samples were collected by fully emptying the breast by electric pump. Table 5 -2 summarizes the concentrations of the measured factors. No leukocytes were detected. Concentrations of total IgA and sIgA, lactoferrin, and lysozyme were similar to those 7 to 12 months postpartum and during gradual weaning. sIgA antibodies to E. coli were produced in the second year, demonstrating significant immunologic benefit to the infant with continued breastfeeding. 89 IgA, IgM, and IgG were measured in nursing women from the beginning of lactation and simultaneously in the feces of their children by Jatsyk et al 130 at the Academy of Medicine in Moscow. They reported IgA to be very high in the milk and rapidly increasing in the feces. IgG and IgM levels, however, were low in both milk and feces. In normal full-term bottle-fed infants, IgA appeared in the feces at 3 to 4 weeks of age but at much lower levels than in breastfed infants. Koutras 146 reported that in the first 8 weeks of life increased amounts of sIgA are found in the stools of breastfed infants compared with formula-fed infants. The authors ascribed this phenomenon to the presence of sIgA in human milk and a stimulation of the local GI production of immunoglobulin.

Savilahti et al 242 measured serum levels of IgG, IgA, and IgM in 198 infants at 2, 4, 6, 9, and 12 months of age. By 9 months, the exclusively breastfed infants had IgG and IgM levels significantly lower than those who had been weaned early (before 3.5 months) to formula. Six infants were still exclusively breastfed at 12 months, and their IgA levels had also lowered to levels found at 2 months with bottle feeders. Infection rates were similar. Two months after the children were weaned to formula, the IgG and IgM levels were comparable. Iron and zinc levels were the same in all children.

sIgA antibodies have been identified in human milk that recognize a large variety of microorganisms. The sIgA antibodies that recognize bacteria, viruses, parasites and fungi are listed in Table 5 85, 196 The list of viruses for which sIgA antibodies exist in human milk is equally long including enteroviruses (poliovirus, coxsackie, and echovirus), cytomegalovirus (CMV), herpes simplex virus, human immunodeficiency virus (HIV), Semliki Forest virus, respiratory syncytial virus (RSV), rubella, reovirus type 3, rotavirus, measles, Norovirus, and porcine coronavirus. IgG and IgM antibodies also exist in human milk against CMV, RSV, and rubella as well as IgE antibodies against parvovirus B19. Noguera-Obenza and Cleary 196 reviewed the role of breast milk sIgA in providing protection for infants against various agents specifically causing bacterial enteritis.

Preservation of human milk at −20° C for up to 3 months does not decrease significantly the levels of IgA, IgG, IgM, C3, C4, lactoferrin, or lysozyme. 65, 70, 160, 205 The preservation of sIgA, IL-6 and TNF-α with freezing at 4° C or −20° C was recently confirmed by Hines et al. 117 A variety of different heat treatments have been applied to milk to protect against bacterial contamination or to protect against infection with specific infectious agents (especially HIV and CMV). Heat treatments include low-temperature, short-time 56° C for 15 minutes; Holder pasteurization 62.5° C for 30 minutes; high-temperature, short-time 70° to 73° C for 15 seconds; boiling 100° C for greater than 1 minute; sterilization, variable time periods, Pretoria pasteurization 56° to 62.5° C for approximately 15 minutes 131 ; flash heating 56° C for approximately 6 minutes with a peak temperature at 72° C 127, 128 ; and microwave heating, with milk temperatures of 20° to 77° C for 30 seconds. 225 Boiling or sterilization essentially destroys 100% of immunologic activity. sIgA and lysozyme activities drop by 20% with Holder pasteurization and by 50% at 65° C. Neither low-temperature, short-time nor high-temperature, short-time reduces the sIgA or lysozyme content markedly. IgG and IgM are greatly reduced by Holder pasteurization. sIgA differs antigenically from serum IgA. IgA can be synthesized in the nonlactating as well as in the lactating breast. It is a compact molecule and resistant to proteolytic enzymes of the intestinal tract and the low pH of the stomach. sIgA present in human milk is primarily manufactured by plasma cells in the mammary gland, modified in its translocation across the mammary epithelia and only minimally produced by the cellular lymphocytes in milk. Levels in milk are 10 to 100 times higher than in serum. Levels in cow milk are very low, that is, a tenth of the level in mature human milk (0.03 mg/dL). Later in life, the human intestinal tract's subepithelial plasma cells secrete IgA. The intestinal secretion of sIgA does not occur in the neonatal period but increases between 4 to 12 months of life.

Discussion continues as to whether any antibodies are absorbed from the intestinal tract, although probably 10% are absorbed. Almost 75% of ingested IgA from milk survives passage through the intestinal tract and is excreted in the feces. All immunoglobulin classes have been identified in the feces. 230 A large body of evidence demonstrates the activity of the immunoglobulins, especially IgA, at the mucosal level of the GI and respiratory tracts. These antibodies provide local intestinal protection against microorganisms, which may infect the mucosa or enter the body through the gut or respiratory tract.

It is well established that the predominant bacteria found in breastfed infants are bifid bacteria. Bifid bacteria are gram-positive, nonmotile, anaerobic bacilli. Many observers have shown the striking difference between the flora of the guts of breastfed and bottle-fed infants. György 100 demonstrated the presence of a specific factor in colostrum and milk that supported the growth of Lactobacillus bifidus. Bifidus factor has been characterized as a dialyzable, nitrogen-containing carbohydrate that contains no amino acid.

In vitro studies by Beerens et al 17 showed the presence of a specific growth factor for Bifidobacterium bifidum in human milk, which they called BB. Other milks, including cow milk, sheep milk, pig milk, and infant formulas, did not promote the growth of this species but did show some activity supporting B. infantis and B. longum. This growth factor was found to be stable when the milk was frozen, heated, freeze-dried, and stored for 3 months. Growth-promoting factors were present for the six strains studied, which varied in their resistance to physical change. Because all these factors were active in vitro, they did not require the presence of intestinal enzymes for activation. It has not been possible to show the presence of this growth factor in other mammalian milks; thus it may contribute to the implantation and persistence of B. bifidum in a breastfed infant' s intestine.

Lactobacillus has been described as one of a number of probiotic bacteria, which provide an immune protective benefit to their host. Lactobacillus reportedly stimulates antibody production and improves phagocytosis by blood leukocytes. 135, 215 The use of probiotic bacteria has reportedly produced benefits in a variety of situations associated with infections. The addition of such bacteria to formula is another example of trying to make formula better by making it more like breast milk. Hatakka et al 112 examined the possible effect of adding probiotic bacteria to formula on the occurrence of infection in children attending day care. They reported modest reductions in the number of children with complicated respiratory infections or lower respiratory tract infections as well as the number of children receiving antibiotics for a respiratory infection in the group of children receiving formula supplemented with Lactobacillus rhamnosus GG compared with children receiving unsupplemented formula.

It was well known in the preantibiotic era that human milk protects human infants throughout lactation against staphylococcal infection. György 100 identified the presence of an "antistaphylococcal factor" in experiments with young mice that had been stressed with staphylococci. This factor, with no demonstrable direct antibiotic properties, was termed resistance factor and described as nondialyzable, thermostable, and part of the free fatty acid part of the phosphide fraction, probably C18:2, but distinct from linoleic acid.

Human milk contains a nonspecific antimicrobial factor, lysozyme, which is a thermostable, acidstable enzyme. This enzyme is a 130-amino-acidcontaining glycoprotein that can hydrolyze the 1-4 linkage between N-acetylglucosamine and N-acetylmuramic acid in bacterial cell walls. It is found in large concentrations in the stools of breastfed infants and not in stools of formula-fed infants; thus it is thought to influence the flora of the intestinal tract.

Goldman et al 92 describe an initial fall in lysozyme levels from 85 to 90 mg/mL to 25 mg/mL at 2 to 4 weeks and then an increase during 6 months to 250 mg/mL (see . Lysozyme levels show an increase over time during lactation; this finding is more apparent in Indian women than in those of the Western world. Reddy et al 229 studied the levels of lysozyme in well-nourished and poorly nourished women in India and found no difference between them (Table 5-4) . As shown in this study, lysozyme levels increase during lactation. Levels in human milk are 300 times the level in cow milk. Lysozyme is bacteriostatic against Enterobacteriaceae and gram-positive bacteria. 219 It is secreted by neutrophils and some macrophages and is present in many body secretions in the adult.

In a study of immunologic components in human milk in the second year of lactation, Goldman et al 89 reported that concentrations of lysozyme, lactoferrin, and total and sIgA were similar to those in uninterrupted lactation and in gradual weaning at 6 to 9 months. sIgA antibodies to E. coli were also produced during the second year. The authors state that "this supports the idea that the enteromammary lymphocyte traffic pathway, which leads to the development of lymphoid cells in the mammary gland that produce IgA antibodies to enteric organisms, operates throughout lactation." 89 When cow milk formula is added to human milk, it reduces the effect of lysozyme; however, powdered human milk fortifier (Enfamil) did not inhibit the antiinfective properties. 141

Lactoferrin is an iron-binding protein closely related to the serum iron transport protein, transferrin, and is part of the larger transferrin protein family. Lactoferrin is found in mucosal secretions (tears, saliva, vaginal fluids, urine, nasal and bronchial secretions, bile, GI fluids) and notably in milk and colostrum. A bacteriostatic effect of lactoferrin is well established for a wide range of microorganisms, including gram-positive and gram-negative aerobes, anaerobes, viruses, parasites, and fungi. The original proposed mechanism of action for its bacteriostatic effect was depriving the microorganism of iron. A second antibacterial action involving direct action with bacterial surfaces; binding negatively charged molecules (lipoteichoic acid) on the surface of gram-positive bacteria neutralizing the surface charge allowing the action of other antibacterial factors like lysozyme or binding lipid A on gram-negative bacteria, releasing the lipid, producing damage to the cell membrane. Another antibacterial action is binding bacterial adhesions blocking host cell interaction. 94 Lactoferrin can kill Candida albicans and C. krusei by changing the permeability of the fungal cell surface. Lactoferrin now is considered a multifunctional, immunoregulatory protein.

The biologic role of lactoferrin has been reviewed in several studies. 163, 164, 198, 241 They point out that lactoferrin reversibly binds two ferric ions and that its affinity for iron is 300 times greater than that of transferrin, retaining iron down to a pH of 3. Human lactoferrin is strongly basic. Lactoferrin is normally unsaturated with iron, 35 and it is usually less than 10% saturated with iron in human milk. 72, 241 Oral iron therapy for an infant can interfere with the bacteriostatic action of lactoferrin, which depends on its unsaturated state for some portion of its bacteriostatic function. Reddy et al 229 showed that giving iron to the mother did not interfere with the saturation of lactoferrin in the milk or thus its potential bacteriostatic effect. Protein energy malnutrition, rather than iron supplies, influences lactoferrin synthesis in the mammary gland. Malnourished but non-iron-deficient mothers are lactoferrin deficient.

The concentration of lactoferrin is high in colostrum-600 mg/dL-then progressively declines over the next 5 months of lactation, leveling at about 180 mg/dL. Breast milk also contains small amounts of transferrin (10 to 15 mg/mL). Lactoferrin is 10% to 15% of the total protein content of human milk. 163 Lactoferrin is resistant to proteolysis, especially in its iron-saturated form. Intact lactoferrin is detectable in the stool of infants, with higher proportions of lactoferrin measurable in the stool of premature infants. 56 Both intact lactoferrin and fragments have been detected in the urine of premature infants, although absorption is less likely in full-term infants. 107 The absorption of iron from breast milk is directly enhanced by lactoferrin. 164 Many bacteria require iron for normal growth, and one bacteriostatic effect of lactoferrin has been ascribed to its iron-binding action. In neutrophils, lactoferrin within neutrophilic granules tightly binds iron, but neutrophils with excessive iron are inefficient at destroying bacteria. Lactoferrin does not limit the growth of all microorganisms; Helicobacter pylori and Neisseria, Treponema, and Shigella species all have receptors for lactoferrin, directly binding iron and allowing adequate growth.

Some evidence supports various other proposed mechanisms of action for lactoferrin' s antimicrobial effect. Lactoferrin has been shown to limit the formation of biofilms by specific organisms, inhibit adhesion to host cells by other organisms and to directly bind to viral particles of herpes simplex virus, HIV, and adenovirus. A proteolytic action of lactoferrin appears to inactivate virulence factors of some organisms. Separately, lactoferrin binds directly to glycosaminoglycans (GAGs) and integrins interrupting the binding of various viruses (herpes simplex virus, HIV, adenovirus, CMV, hepatitis B virus [HBV]) to host cells. Pepsin hydrolysate products of lactoferrin (B or H) may exert a direct bactericidal effect by binding to lipopolysaccharide of gram-negative organisms and disrupting bacterial membranes. 263 Lactoferrin may cause an increased release of cytokines by cells including antiinflammatory cytokines such as IL-10. 50, 157 Others have shown that lactoferrin suppresses the release of IL-1, IL-2, IL-6, IL-8, and TNF-α, all proinflammatory cytokines, which would be more of an immune-modulating effect. 157 Other investigators using a recombinant human lactoferrin (talactoferrin) demonstrated evidence of lactoferrin causing increased maturation of DCs 252 and talactoferrin causing the recruitment and activation of neutrophils and macrophages 233 as other examples of how lactoferrin affects the innate immune protection of the growing infant. Several other effects have been proposed for lactoferrin, including inhibition of hydroxyl radical formation, decreasing local cell damage; lipopolysaccharide binding, also leading to a diminished inflammatory response; and DNA binding, affecting transcription and possibly regulation of the production of cell products. 198 Activation of natural killer (NK) cells, modulation of complement activity, and blocking of adhesion of enterotoxigenic E. coli and Shigella flexneri 98 are other proposed actions of lactoferrin. A specific region of lactoferrin, near the N terminus of the molecule, is strongly basic and is reported to mediate some of lactoferrin' s antimicrobial activity. "Lactoferricins," small peptides containing this basic region, produced by proteolytic cleavage reportedly bind to lipopolysaccharide, leading to disruption of the bacterial cell wall and cytoplasmic membrane. 263 In another area of immune protection, lactoferrin may limit cancer development. 157 The proposed mechanisms of its anticancer effects includes increasing NK cell cytotoxicity, increased production of IL-18 and inhibition of angiogenesis, augmented apoptosis of cancer cells and initiation of cell cycle arrest in growing tumor cells. 157 The multiple roles and proposed mechanisms of action of lactoferrin in breastfed infants continue to be more specifically elucidated.

Colostral cells in culture have been shown to be stimulated to secrete an interferon-like substance with strong antiviral activity up to 150 National Institutes of Health units/mL. 219 This property has not yet been identified in the supernatant of colostrum or milk. Interferon-γ has been produced by T cells from human milk when stimulated in vitro. 219 The T cells isolated from human milk were the CD45RO phenotype and have been identified as a source of interferon. Srivastava et al 254 have measured low levels of interferon-γ in not only colostrum but also transitional and mature milk. They postulated that the low level of interferon-γ (0.7 to 2 pg/mL) might be adequate to protect against infection without hyperactivation of T cells. Interferon is produced by NK cells and by T cells, phenotypically Thy0 and Thy1. It can cause increased expression of major histocompatibility complex molecules, increase macrophage function, inhibit IgE and IL-10 production, and produce antitumor and antiviral activity. The exact role of interferon-γ in breast milk has not been delineated.

The C3 and C4 components of complement, known for their ability to fuse bacteria bound to a specific antibody, are present in colostrum in low concentrations compared with their levels in serum. IgG and IgM activate complement. C3 proactivator has been described, and IgA and IgE have been identified as stimulating the system. Activated C3 has opsonic, anaphylactic, and chemotactic properties and is important for the lysis of bacteria bound to a specific antibody. No functional role for complement in breast milk has been identified.

Unsaturated vitamin B 12 -binding protein of high molecular weight has been found in very high levels in human milk and in the meconium and stools of breastfed infants compared with its levels in infant formulas and infants who are formula fed. The protein binding renders the vitamin B 12 unavailable for bacterial growth of E. coli and Bacteroides. 99

Glycans are complex carbohydrate structures attached to various other structures (a lactose moiety, a lipid component, peptides, proteins, or aminoglycans) that are present in large amounts in human milk. 192 They include glycoproteins, glycolipids (gangliosides), glycosaminoglycans, mucins, and oligosaccharides. Oligosaccharides are composed of a basic core structure derived from glucose, galactose, or N-acetylglucosamine and are linked to a variety of terminal fucose linkages or sialic acid linkages to create numerous different compounds. Oligosaccharides compose the major portion of glycoconjugates in milk and are present in the milk-fat globule membrane and in skim milk. 188, 192 Gangliosides are glycolipids found in the plasma membrane of cells, especially in cells in the gray matter of the brain. More specifically, gangliosides are glycosphingolipids that contain sialic acid, hexoses, or hexose amines as the carbohydrate component and ceramide as the lipid component of the molecule. The predominant gangliosides in human milk are GM1, GM2, GM3, and GD3, as reported by Newburg. 190 A diverse abundance of these complex carbohydrates are synthesized by the many glycosyltransferases contained in the mammary gland. Mucin and lactadherin are two glycoproteins included in this group that have antimicrobial effects. 238 Some of these carbohydrate molecules are structurally similar to glycans on the surface of small intestine epithelial cells that act as receptors for microorganisms. One proposed mechanism for the antimicrobial effect of these soluble substances is direct binding with the potential pathogenic organisms. 189, 190 Schroten et al 247 proposed that mucins contained in the human milk fat-globule membrane can block bacterial adhesion throughout the intestine after studying the adhesion of S-fimbriated E. coli to buccal epithelial cells.

Gangliosides appear to be responsible for blocking the activity of heat-labile enterotoxin from E. coli and the toxin from Vibrio cholerae in rat intestinal loop preparations. 207 Another toxin from Campylobacter jejuni, with similar binding specificity, also seems to be inhibited by GM1. 151, 235 Globotriaosylceramide, another glycolipid in human milk, is the natural cell surface receptor for the toxin from Shigella dysenteriae and verotoxin released by enterohemorrhagic E. coli. 191 The proposed mechanism of action of these glycolipids is that by binding to the toxin they form a stable complex that prevents the toxin from binding to the appropriate receptors on intestinal cells; however, Crane et al 48 proposed from their studies that the oligosaccharide binds to the toxin receptor to block the action of the heatstable enterotoxin of E. coli. Human milk gangliosides may be important in protecting infants against toxin-induced diarrhea, but this has not been specifically demonstrated in vivo in controlled trials. 191, 207 Evidence exists that human milk glycans inhibit a broad range of pathogens (Table 5-5) . [188] [189] [190] [191] [192] [193] Newberg et al 191 document the constitutive expression of various fucosylated glycans in human milk and secretions and present "typical" concentrations of these active agents in human milk from the literature. Their secretion is related to the "secretor" and Lewis genes, which control the individual differences in expression of Lewis blood group types.

Chaturvedi et al 44 have recently examined the survival of oligosaccharides from human milk in infants' intestines. They demonstrated that the concentrations of oligosaccharides were higher in the infants' feces than in mothers' milk and higher in feces than urine. The profile of oligosaccharides found in the infants was similar to that found in their mothers' milk. The formula-fed infants had lower concentrations of oligosaccharides and the profiles of the oligosaccharides were different from those found in the breastfed infants. The oligosaccharides remained intact passing through the intestine. A small percentage are absorbed and excreted intact in the urine. The oligosaccharides were available at these sites to block intestinal and urinary pathogens. Two other groups of researchers have documented variation of the composition of glycans in human milk over the first 4 months of lactation 47 and variations in the composition of glycans in diverse populations. 64 Therefore a diverse repertoire of glycans are present in large amounts in human milk, which persist intact in the intestine and reach the urine, and have demonstrated inhibitory effects on a variety of pathogens. These components constitute a major contribution of human milk to innate immunity at the level of an infant' s gut.

Ganglioside GM 1 Cholera toxin Other authors propose that the gangliosides GD3 and GM3 may play an immunomodulatory role early in lactation by affecting DCs decreasing the production of interleukins (IL-10 and IL-12) and suppressing the expression of various cluster designation (CD) markers and major histocompatability complex class II on DCs. 26 

Interleukins (ILs) are considered a "subgroup" of cytokines. 161 Originally, when cytokines were first hypothesized, it was thought that they were primarily produced by leukocytes and acted on other leukocytes, and therefore they could be called ILs. Although much of their effect is on lymphocyte activation and differentiation, it is now known that ILs act on and are produced by a variety of cells. 91 Goldman et al 91 identified IL-1β, IL-6, IL-8, and IL-10 in breast milk (Table 5-6). Srivastava et al 254 reported measuring moderate amounts of IL-6, IL-8, and IL-10 in the different stages of breast milk. Very low amounts of IL-1β were detected, especially in comparison with the amount of IL-1 receptor antagonist (RA), which presumably could block the activity of the small amount of IL-1. Hawkes et al 113 reported on the amount of cytokines in breast milk over the first 12 weeks of lactation. The proposed "proinflammatory" cytokines, IL-1β, IL-6, and TNF-α, were present in only 7 of 36 mothers who donated samples at each point throughout the study. A broad range of concentrations of each of these cytokines was seen during the course of the study. The "antiinflammatory" cytokines, transforming growth factor (TGF) -α 1 and TGF-β 2 , were present in significant amounts in all samples. IL-2 has also been reported in breast milk in 81% of the mothers tested, with milk (aqueous) levels correlating with plasma IL-2 levels. IL-2 was constitutively produced from 57% of milk cell samples and IL-2 production was markedly increased by stimulation of the cells with Con A. 29 IL-6 has been identified in breast milk by other investigators, especially in the first 2 days of life. 209, 234 The authors suggest that IL-6 in human milk may augment the newborn' s immune functions before the body can begin full production of cytokines. Specifically, this is accomplished by increasing antibody production, especially IgA; enhancing phagocytosis; activating T cells; and increasing α 1 -antitrypsin production by mononuclear phagocytes. IL-7 is a chemokine known to improve thymic output in animals and appears related to the proliferation and survival of T cells in all stages of development. 194 Ngom et al 194 have described improved thymic function in exclusively breastfed infants associated with higher IL-7 concentrations in the mother' s breast milk. The breast milk of Gambian mothers contained variable levels of IL-7, but the geometric mean levels were higher in the first 8 weeks postpartum in mothers whose infants were born in the "harvest-season" (January to June) compared with those mothers whose infants were born in the "hungry-season." The authors postulate that IL-7 in breast milk enhances T-cell proliferation and survival and overall thymic development in the infant, leading to long-term benefits in protection from infection.

IL-8 is a chemokine capable of attracting and activating neutrophils and attracting CD45RA + T cells. IL-8 is produced by mammary epithelial cells. 209 Srivastava et al 254 also detected messenger ribonucleic acid (mRNA) for IL-8, suggesting that cells in breast milk were capable of producing IL-8. The exact function of IL-8 in breast milk remains to be elucidated.

IL-10 is thought to have antiinflammatory effects, including decreasing the production of interferon-γ, IL-12, and other proinflammatory cytokines. It has been reported to enhance IgA, IgG, and IgM synthesis. IL-18 has been identified in colostrum, early milk, and mature milk with the highest levels occurring in colostrum and in association with preterm deliveries and complications of pregnancy in the mothers. 260 The levels of IL-18 were correlated with soluble Fas ligand in colostrum. IL-18 was detected by immunohistochemical staining in actively secreting epithelial cells in a lactating breast. IL-18 has been shown to be produced by intestinal epithelial cells and activated macrophages. It leads to the production of other chemokines (GM-CSF, IL-2, TNF-α). It induces the expression of Fas ligand on lymphocytes. The authors suggested that IL-18 present in colostrum may play a role in stimulating a systemic T H 1 response and causing NK cell and macrophage activation in neonates.

The interaction and the direct effect of these ILs in breast milk must be clarified. The amount of T cells bearing markers of recent activation is increased in human milk compared with the results in peripheral blood of adults. Wirt et al 276 

Of the many bioactive substances that have been identified in human milk, cytokines are some of the most recently identified and investigated agents, although their existence has long been suspected in attempts to explain certain immunologic and protective effects of breast milk on infants. More than 40 cytokines have been described, 166 and more than 10 of these have been identified in human milk. 91, 254 Cytokines are small proteins or glycoproteins that, through binding to receptors on immune and nonimmune cells, produce a broad range of effects (many still unidentified) through autocrine, paracrine, and endocrine actions. Cytokines are produced predominantly by immune cells and function in complex associations with other cytokines to stimulate and control the development and normal functioning of the immune system. The nomenclature and abbreviations used are complicated and confusing. Newer systems of classification have been established according to which cells produce them or what their general functions are 161 or based on the relative position of their cysteine residues or their receptor types (CCR, CXCR, CX3CR). 166 Box 5-3 provides a simplified list with abbreviations.

Little evidence demonstrates specific in vivo activity of the different cytokines. Based on general information on the function and interaction of the particular cytokines, as well as consideration of as yet unexplained effects of breast milk, proposed functions of the cytokines include initiation of development of host defense, stimulation of host defenses, prevention of autoimmunity, antiinflammatory effects in the upper respiratory tract and GI tract, and stimulation of the development of the digestive system, especially the mucosal immune system of the alimentary tract and the proximal respiratory tract. The maternal breast may respond to feedback stimulation or suppression by secreted cytokines, influencing the growth, differentiation, and secretory function of the breast. As shown in other situations, cytokines may enhance receptor expression on cells in the respiratory and GI tracts for major histocompatibility complex molecules or immunoglobulins. Various cell types in the mucosal immune system may be activated or attracted to specific sites in the GI tract by the action of cytokines.

Beyond these proposed beneficial effects of cytokines, newer studies are identifying specific Leukocyte inhibitory factor LIF immunologic and protective roles for different cytokines in developing infants. For example, extensive work has been done on epidermal growth factor (EGF) and other growth factors (HB-EGF, G-CSF, EPO, and EPO-like growth factors) have been studied relative to their role in preventing necrotizing enterocolitis (NEC) and gut homeostasis. 183 A number of potential roles for EGF in gut homeostasis have been proposed and studied, including intestinal development, proliferation and adaptive response to damage, repair and regeneration and diminishing inflammatory responses to various stimuli. TGF-β has been studied for its role in initiating and stimulating IgA production early on in infancy. 200 The actual measurement of cytokines in breast milk has been complicated by a number of factors, including different assays used (bioassays, enzyme-linked immunosorbent assay [ELISA], radioimmunoassay), binding to proteins, their existence in monomeric or polymeric forms, 8 the presence of antagonists, and their varying presence in colostrum, early milk, or mature milk. Goldman et al 91 reported on the bioactivity and concentration of cytokines in breast milk from their own work and that of others (see Table  5 and at lower levels in mature milk, as well as high levels of soluble TNF-α receptor I (sTNF-αRI), sTNF-αRII, and IL-1RA. Also, they identified that most TNF-α did not exist "free" in breast milk but was associated with TNF receptors. The in vivo significance of these findings remains to be assessed.

Given the complex interaction and regulation of cytokine production and cytokines' relation to coordinated inflammatory and antiinflammatory responses in tissues, one should assume that the interaction of cytokines in breast milk and the effect of cytokines, cytokine receptors (soluble and expressed on various cell types), and cytokine antagonists on the infant will be equally complex. A new methodology, antibody-based protein arrays has been applied to identify cytokines in human milk. 148 Kverka et al 148 analyzed colostrums and milk samples from the first 4 days postpartum using two different arrays capable of detecting 42 and 79 cytokines. Three cytokines (EGF, IL-8/CXCL8, GRO/CXCL1-3) were detected in all of the tested samples. Nineteen cytokines were present in more than 50% of the samples. An additional 32 cytokines were identified in human milk for the first time. The concentration of cytokines varied in the different women and varied over time. Continued investigation with this and other assays will be essential to understanding the significance and specific effects of these substances in breast milk.

Nucleotides, nucleosides, nucleic acids, and related metabolic products are essential to many biologic processes. Although they are not essential nutrients because they can be synthesized endogenously and recovered from in vivo "salvage" sources, their presence in the diet may carry significant benefits under various conditions (i.e., "conditionally essential"). 42, 184, 266 In situations of disease, stress, rapid growth, or limited dietary intake, supplementation of the diet with nucleotides may decrease energy expenditure to synthesize or salvage nucleotides, which optimizes the host response to these adverse situations.

Nucleotides exist in relatively large amounts in human milk, 15% to 20% of the nonprotein nitrogen, suggesting that they have some nutritional significance, although no clinical syndromes have been associated with nucleotide deficiency to date. Nucleotides are present in the natural milk of different species in varying amounts and composition. The nucleotide content and composition of bovine milk are particularly less and different from human milk. Infant formulas supplemented with nucleotides contain roughly the same amounts of nucleotides as human milk, from 20 to 70 mg/L. 39, 40, 154 Unsupplemented formulas contain less and different amounts of nucleotides.

Mammalian cells contain a large variety of nucleotides and related products, which have many metabolic functions, including the following 40-42 :

1. Energy metabolism: adenosine triphosphate is a major form of available cellular energy. 2. Nucleic acid precursors: the monomeric units for RNA and DNA are present. 3. Physiologic mediators: cyclic adenosine monophosphate and cyclic guanosine monophosphate serve as "messengers" for cellular processes; adenosine diphosphate is necessary for platelet aggregation; and adenosine has been shown to affect vasodilatation. 4. Related products function as coenzymes in metabolic pathways: nicotinamide-adenine dinucleotide, flavin adenine dinucleotide, and coenzyme A.

rying molecules in synthetic reactions: uridine diphosphate glucose in glycogen synthesis and guanosine diphosphate mannose, guanosine diphosphate-fucose, uridine diphosphate-galactose, and cytidine monophosphate sialic acid in glycoprotein synthesis. 6. Allosteric effectors: the intracellular concentrations of nucleotides influence the progression of certain steps of metabolic pathways. 7. Cellular agonists: extracellular nucleotides influence intracellular signal transduction (e.g., cyclic adenosine monophosphate and inositol-calcium pathway).

Nucleotide concentrations in cells and tissues are maintained by de novo synthesis and salvage from intermediary metabolism and diet ( Figure 5-11 ). 224 Nucleosides are the predominant product absorbed in the small intestine. Nucleosides are probably transported by passive diffusion and a carrier-mediated process; purines and pyrimidines are transported by passive diffusion at high concentrations and by a sodium-dependent active mechanism at low concentrations ( Figure 5-12) . 224 The digestion and absorption of nucleotides, nucleosides, and pyrimidines and purines also involve polymeric and monomeric nucleotides and other adducts (nucleosides in a biologically active moiety).

In early reports on the nucleotide and nucleoside content of milk, various methods of measurement were used, and the amounts were described as either the monomeric fraction of nucleotides or the total RNA. Leach et al, 154 recognizing the complex nature of digestion and absorption of nucleotides and related products, attempted to measure the total potentially available nucleosides (TPANs) in human milk using solid-phase extraction, highperformance liquid chromatography analysis, and enzymatic hydrolysis of the various fractions. They analyzed breast milk samples at various stages throughout lactation (colostrum; transitional, early, and late mature milk) from 100 European women and 11 American women. They used an aqueous TPAN-fortified solution containing ribonucleosides, 5′-mononucleotides, polymeric RNA, and nucleoside-containing adducts to estimate the accuracy of their process.

The mean ranges of TPAN values were similar for European women from different countries and American women, although broad ranges were seen and the composition of individual nucleotides varied. 154 The mean TPAN value was lowest in colostrum but did not show a consistent upward or downward trend in transitional, early, or late mature milk. The mean ranges of TPAN values were 82 to 164 mmol/L for colostrum, 144 to 210 mmol/L for transitional milk, 172 to 402 mmol/L for early mature milk, and 156 to 259 mmol/L for late mature milk (Table 5-7) . Monomeric and polymeric nucleotides were the predominant forms of TPAN in pooled samples. Cytidine, guanosine, and adenosine were found mainly in these fractions, whereas uridine was found primarily as free nucleotide and adduct ( 154 concluded that their process of estimating TPANs, including sequential enzymatic hydrolyses, and measuring the entire nucleotide fraction provides a reasonable estimate of the in vivo process and the nucleotides available to the infant from human milk.

Proposed effects of dietary nucleotides include effects on the immune system, iron absorption, intestinal flora, plasma lipoproteins, and growth of intestinal and hepatic cells. Effects on the immune system, related to nucleotide supplementation to the diet, have mainly been reported from animal studies and include increased mortality rate from graft-versus-host disease, improved delayed-type cutaneous hypersensitivity and alloantigen-induced lymphoproliferation, reversal of malnutrition and starvation-induced immunosuppression, increased resistance to challenge with S. aureus and C. albicans, and enhanced T-cell maturation and function. 218 Spleen cells of mice fed a nucleotide-free diet produce lower levels of IL-2, express lower levels of IL-2 receptors, and have decreased NK cell activity and macrophage activity. 40, 41 Presumably these nucleotide-associated changes are related to T-helper/inducer cells and the initial phases of antigen processing and lymphocyte proliferation. 41, 42, 266 In vitro and in vivo experiments have documented that ingested nucleotides increased iron absorption, perhaps affecting xanthine oxidase. 224 Although in vitro studies showed that added nucleotides enhanced the growth of bifidobacteria, conflicting results have been obtained on the influence of dietary nucleotides on the fecal flora of infants receiving breast milk or nucleotide-supplemented formula. 11, 224 Clinical studies in infants receiving nucleotide-supplemented formula demonstrated increased high-density lipoprotein cholesterol, lower very-low-density lipoprotein cholesterol, increased long-chain polyunsaturated fatty acids, and changes in red blood cell membrane phospholipid composition. 11 Supplementation studies in animals have shown enhanced GI tract growth and maturation, improved intestinal repair after diarrhea, stimulation of hepatic growth, and augmented recovery from hepatectomy. 218 A recent review discusses the effects of dietary nucleotides on the immune system and protection against infection reported in studies in the literature. 244 Carver et al 41 †Adducts are of the form nucleoside-phosphate-phosphate-X, where X is a biologically relevant moiety (e.g., uridine diphosphate-galactose or nicotinamide-adenine dinucleotide). 4 months, NK cell activity and IL-2 production were higher in the breastfed and nucleotidesupplemented groups compared with those receiving formula without nucleotide supplements. Infections occurred infrequently in all groups, but slightly less in the breastfed group. No differences were noted in hematologic profiles and plasma chemistry values, and no toxicity or intolerance was associated with nucleotide supplementation. The sample size was small, marked variability was seen in the IL-2 measurements, and the differences noted at 4 months were less than at 2 months. Therefore the authors concluded that dietary nucleotides may contribute to improved immunity in breastfed infants.

Brunser et al 28 examined the effect of a nucleotide-supplemented formula on the incidence of diarrhea in 392 infants in Chile, studied through 6 months of age. Although the infants receiving the supplemented formula (20 mg/L) experienced less diarrhea, the difference in the duration of diarrhea was small. The numbers were too small to comment on the causative agents of diarrhea, although no apparent protection against any one agent was seen. The beneficial effect of nucleotides against diarrhea was proposed to be secondary to enhanced immune response to intestinal pathogens or improved intestinal integrity or a combination of both. In a larger study of 3243 infants younger than 6 months of age, the severity of the diarrhea (duration and number of bowel movements) as well as the incidence of diarrhea was lower in the nucleotide supplemented group. 152 Two groups of premature infants fed either nucleotide supplemented (20 mg/L) or unsupplemented formula were followed, measuring the concentration of plasma immunoglobulins throughout the first 3 months of life. 184 IgG plasma concentrations were not different in the two groups during the study period. IgM plasma levels were higher in the nucleotide supplemented group at 20 to 30 days and 3 months of life, while IgA plasma levels were significantly higher at 3 months of age in the supplemented group.

Pickering et al 218 published a 12-month, controlled, randomized study of 311 infants to examine the effect of added nucleotides at levels comparable to human milk on infants' immune responses to various vaccine antigens; 103 nonrandomized infants received breast milk for at least 2 months and then either human milk or a standard infant formula. Another 208 infants were randomized to receive either a standard infant formula or one supplemented with nucleotides. The amount and actual nucleotide content added were based on TPANs, as measured by Leach et al, 154 equaling 72 mg/L. Overall growth and nutrition tolerance were similar in each group. The nucleotide group had significantly higher geometric mean titers of H. influenzae type b antibody and diphtheria antibody than the control group or the breastfed infants. No significant difference was seen between the nucleotide and control groups for the IgG response to oral poliovirus vaccine or tetanus. Infants who are breastfed for longer than 6 months had significantly higher antibody responses to oral poliovirus vaccine than children breastfed for less than 6 months or either of the two formulafed groups. No significant differences were found between the different groups with respect to total IgG, IgA, or IgE. Differences were seen in the number of children who experienced at least one episode of diarrhea: the nucleotide group (4/27, 15%) versus the control group (13 of 32, 41%, p <0.05) and the breastfed group (6 of 27, 22%) . Notably, the breastfed group was heterogeneous relative to the amount of breast milk received and the duration of feeding, whereas the nucleotide group received supplementation for the entire 12 months.

Questions that remain concerning nucleotides and their proposed beneficial effects in an infant' s diet include the following:

• What are the proven mechanisms of action of these proposed benefits? • What form and concentration of nucleotides are necessary to effect these benefits? • Is adequate information available to justify using nucleotides in infant formula in higher amounts and different compositions than are currently used?

Debate and research to answer these and other questions concerning nucleotides will continue.

A primary function of each of the body' s different mucosal surfaces is immunologic. Each distinct mucosal surface has multiple other physiologic functions including gas exchange (in the lungs), nutrient absorption (in the gut), sensory detection (in the eyes, nose and mouth) and reproduction (in the uterus and vagina). The thin, permeable nature of these barrier mucosal surfaces, their large surface area and the constant exposure to microorganisms, foreign proteins, and chemicals predisposes the mucosal membranes to damage and infection. During the first year(s) of life, when the infant' s immune system is developing and maturing, it is doing so on a systemic and a mucosal basis as well as involving both innate and adaptive immune mechanisms. That development must include the ability to respond to and protect against invasive pathogens at the same time as "tolerate" or "ignore" the multitude on commensal organisms that reside at these surfaces. During this early development, breast milk contains numerous bioactive factors that supplement the immune protection at the mucosal level while limiting inflammation and contribute to the immune modulation and growth stimulation of infants' mucosal and systemic immune defenses. The mucosal immune system involves both innate mechanisms and adaptive immune mechanisms functioning in concert. The development of the mucosal immune system occurs in the prenatal period and continues in the postnatal period. The functional mucosal barrier includes the action of enzymes, chemicals, acidity or pH, mucus, immune globulins and indigenous flora. In as early as 8 weeks of gestational age, researchers have identified changes in the intestinal barrier with the development of enterocytes, goblet cells and enterochromaffin cells along with evidence of development of tight junctions between the epithelial cells. 211, 222 Mucus production, which can block adherence of pathogens to epithelial cells, demonstrates both pre-and postnatal development beginning with evidence of expression of the muc2 gene as early as 12 weeks' gestational age. 34 This is approximately the same time that Paneth cells appear in intestinal crypts. These cells secrete various products, including α-defensin, lysozyme, secretory phospholipase A 2, and TNF-α, which contribute to protection from pathogens, stem cell protection within the epithelia layer and influence the selection and number of commensal organisms. 180, 240 Secretory immune globulins sIgA and IgM act at the epithelial surface, largely without inflammation, by limiting adherence and transmigration and facilitating phagocytosis of potential pathogens.

The well-recognized MALT is present in localized areas beneath the mucosal surfaces: tonsils and adenoids in the nasopharynx and Peyer patches and isolated lymphoid follicles in the intestine. Overlying the isolated lymphoid follicles of the gut are specialized epithelial cells called M-cells. M-cells (membrane, microfold, or multifenestrated cells) come in direct contact with microorganisms and antigens due to a lack of a surface glycocalyx covering. These remarkable cells endocytose, phagocytose and transcytose molecules, and antigens from their luminal surface to their basal surface. Antigenpresenting cells and lymphocytes process the trancytosed molecules, presenting them to submucosal aggregates of lymphocytes. The activated lymphocytes that have responded to the specific presented antigens migrate via the lymphatics to the thoracic duct and into the blood. These lymphocytes circulate in the blood, until they return to mucosal tissues, predominantly the same ones they originated from, where they now function as effector lymphocytes in the lamina propria. This process of "directed migration" to specific sites occurs due to the influence of cytokines and adhesion molecules, such as chemokine CCL28 (mucosal epithelia chemokine) expressed in the colon and salivary glands and CCL25 (thymus-expressed chemokine) which effects the site-specific migration. 217 The immune response of lymphocytes in the submucosa and the subsequent directed migration to the same and other mucosal sites produces a focused response to a selected repertoire of antigens at those sites. The lactating mammary gland is an essential component of MALT. A mother' s mature effective immune response to microorganisms in her and her infant' s environment through this antigenic stimulation of MALT in the mother' s gut and respiratory mucosa and the directed migration of cells to the breast results in activated lymphocytes and antibodies in the breast milk providing protection to the infant against those microorganisms. This is a wellrecognized example of how breast milk can provide additional immune protection to the infant. It is also one of the reasons to continue breastfeeding when a mother or the infant have a possible infection. The mucosal immune system undergoes significant postnatal development, in part due to the dramatic exposure of the mucosa to large numbers of microorganisms in early postnatal life. Peyer patches are rudimentary, and few immunoglobulinproducing intestinal plasma cells are present until several weeks after birth. 24 After several weeks, germinal centers within the lymphoid follicles develop, and the number of IgM-and IgA-producing cells in the intestine increase. Immunoglobulin-producing intestinal plasma cells (primarily IgA-producing cells) in the lamina propria increase in number from 1 to 12 months of age. 257 With normal maturation of the mucosal immune system, large numbers of immunoglobulin-producing cells locate in the intestinal lamina propria. The monomeric IgA produced by these plasma cells is transported through epithelial cells to the mucosal lumen. Attachment of an epithelial glycoprotein, the membrane secretory component to two IgA molecules leads to the formation of a dimeras; the sIgA molecule is "secreted" at the mucosal surface. IgM, in the form of a pentamer, contains a polypeptide J-chain and is transported by the same mechanism. 25 A portion of the secretory component remains attached to the sIgA and IgM, which protects these molecules against proteolysis and contributes to their stability. Large amounts of sIgA and IgM are produced, in a similar fashion, by the mammary glands and delivered to the infant via breast milk. The sIgA and IgM remain stable in saliva and feces 105 and provide specific protection by blocking adherence and entry and facilitating inactivation, neutralization, and agglutination of a wide variety of microorganisms.

Distinct from the action of immunoglobulins, a large number of bioactive factors in breast milk act at the mucosal level to supplement the innate defenses. 103 These include lactoferrin, lysozyme, casein, oligosaccharides, glycoconjugates and lipids. Mucin-1, lactadherin, and a glycosaminoglycan are antimicrobial components, which are part of the milk-fat globule. Free-fatty acids and monoglycerides, digested components of the milk-fat globule, can cause lysis of enveloped viruses, bacteria, fungi, and protozoa. Lauric and linoleic acids, which constitute a large percentage of the FFE in human milk, are two such acids produced by lipolysis in the stomach. 104 Additional factors contained in breast milk with demonstrated activity at the level of the mucosa include cytokines, hormones and growth factors. IL-10 and IFN-γ act by influencing epithelial barrier integrity. 74 Other factors that are considered to contribute to mucosal growth and development are TGF-α, EGF, and hormones (insulin and insulinlike growth factor). 56 Many other factors contained in breast milk have the potential for activity at the level of the mucosa, including nutrients, vitamins, nucleotides, enzymes, and soluble molecules with receptor-like structures (soluble CD14, soluble toll-like receptor 2) 150,155,267

Toll-like receptors (TLRs) and the complex interaction between indigenous bacterial flora and the intestine is an important aspect of research into the development of the mucosal immune system. Forchielli and Walker 69 have reviewed many of these immune mechanisms acting at the mucosal level. TLRs are transmembrane receptors (pattern recognition receptors) that are capable of detecting and discriminating among various groups of potential pathogens and initiate different immune responses to them. TLRs "recognize" pathogenassociated molecular patterns, conserved features in the pattern of molecules expressed by pathogens and commensal organisms. Specific TLRs recognize a particular repertoire of patterns: TLR2 identifies bacterial lipoproteins and peptidoglycan molecules; TLR3 recognizes double-stranded DNA and TLR4 identifies lipopolysaccharide. Ten TLRs are recognized in humans to date; some have identified legends (pathogen-associated molecular patterns from viruses, bacteria, and protozoa) to which they bind. TLRs are present on some epithelial cells, but are predominantly expressed on macrophages and DCs. 228 Intestinal epithelial cells are influenced by gut flora and local immune response to express specific TLRs. The recognition of specific antigens by epithelial cell TLRs stimulates different intracellular signal pathways that lead to different T-lymphocyte immune responses. It has been postulated that the ongoing immune stimulation elicited by the microbial flora in the gut "programs" the host to predominately express different T-helper cell responses: T H 1-like, T H 2-like and T H 3-like. This is referred to as "crosstalk" between the indigenous intestinal flora and the body' s immune system. The T H 1-like response is described as delayed-type hypersensitivity or cellular immunity; characterized by the predominant release of IL-2, IL-12 and interferon-γ. The T H 2-like is primarily humoral immunity, antibody production (especially IgE) associated with ILs: IL-4, IL-5, and IL-6. The T H 3-like response is related to oral tolerance and antiinflammatory effects in association with the release of IL-10 and TGF-10. A theoretical "ideal" for this system is the ability of the host to respond to various stimuli with balanced protection against the microbial invasion without excessive inflammation or damage to the host. An imbalanced (or poorly regulated) response of this system could result in an allergic reaction against food proteins (T H 2 excess) or an autoimmune inflammatory response against self-antigens (T H 1 excess). 69 Ongoing research continues to explore these molecular mechanisms, their potential contribution to allergy, autoimmune disease and normal immune function development within a fetus, infant, and young child. The role of breast milk in the development of the systemic and mucosal immune systems takes on new significance when considering these concepts and mechanisms. This is especially true when examining the role of breast milk in adding to the innate and adaptive immune response at the level of the mucosa, the effect breast milk has on the respiratory and intestinal microbiota, the purported antiinflammatory effects of breast milk, and the proposed influence of breast milk on the maturational development of the intestine. Vorbach, Capecchi, and Penninger 270 postulated that the mammary gland evolved from a protective immune gland as part of the innate immune system. They present a list of various protective molecules that are part of both mucosal secretions and human milk. They discuss how specific nutritional factors in human milk have dual functions: nutritional and protective. This highlights the dual role of the breast as a nutritional and immune organ and should stimulate further research into the breast' s role in innate immunity as a component of the mucosal immune system.

Investigation into the microbial colonization of the intestinal tract has exploded. Much of this investigation has been driven by new molecular techniques involving the analysis of ribosomal RNA sequences of microbes that might not have been identified by traditional culture techniques. The diversity of the microbiota (all the microbes which colonize the GI tract) can be viewed from different perspectives based on the technical methods used. 255 Probiotics have been broadly defined as microorganisms that can exist within a host while affording benefits for the organisms and the host. Prebiotics are substances that (through different mechanisms) increase the growth and survival of probiotic bacteria within the host. Commonly recognized probiotic bacteria are Lactobacillus rhamnosus GG, Bifidobacteria infantis, Streptococcus thermophilus, Bacillus subtilis, Saccharomyces boulardii, and Bifidobacteria bifidus. Many more organisms are considered to be probiotic, some of which are commercially available. 178 Prebiotics are predominantly nondigestible oligosaccharides that ferment within the colon changing the ambient pH and producing small-chain fatty acids. Breast milk with its significant composition of oligosaccharides functions as a prebiotic source for an infant, facilitating the growth of bifidobacteria and lactobacilli. 52, 249 Ongoing research is exploring the potentially mutually beneficial relationship between the microbes and the host with particular attention to nutrition (the availability of nutrients, energy sources, and synthesis of vitamins as influenced by the microbes), the developing GI tract (including angiogenesis and mucosal barrier repair), 75, 186, 210 the maturation of mucosal immunity, 140,216 both the innate system 119 and adaptive system, 170 and the bioavailability and metabolism of drugs and chemicals in the GI tract. 57, 121 Specific proposed mechanisms of how probiotic bacteria and prebiotic substances contribute to an infant' s developing immune system include competition with pathogenic bacteria for colonization, strengthening the tight junctions to enhance the mucosal barrier, producing antimicrobial bacteriocidins, stimulating mucus production, stimulating peristalsis, influencing the secretion of sIgA, stimulating the crosstalk interaction between intestinal cells and colonizing bacteria to affect the mucosal immune development, and increasing the production of certain cytokines (IL-10 and interfer on-γ). 75, 140, 214, 216 

The microbial colonization of an infant' s intestinal tract begins at birth, with organisms from the maternal flora being the first colonizers. Numerous additional factors directly influence the composition of the intestinal microbiota in early infancy, including gestational age, ingestion of breast milk or formula, initiation of solid foods, mode of delivery, the route of delivery of food, the time of onset of feeding, exposure to other microbes through contact (with family, animals, persons from other environments), antibiotics, and intercurrent or chronic illness. 66, 215 The predominant flora of breastfed infants are Lactobacillus bifidus and Bifidobacterium spp., which constitute up to 95% of the culturable organisms. The remaining minority of bacteria include Streptococcus, Bacteroides, Clostridium, Micrococcus, Enterococcus, E. coli, and other uncommon organisms in small numbers. 178, 186, 282 Lactobacillus bifidus metabolizes milk saccharides, producing large amounts of acetic acid, lactic acid, and some formic and succinic acids, which create the low pH of the stool of breastfed infants. L. bifidus also produces short-chain fatty acids in the course of colonization. Large numbers of bifidobacteria can lower the pH of the intestine, which limits the growth of some pathogens such as E. coli, Bacteroides, and staphylococci. The flora of bifid bacteria is inhibitory to certain pathogenic bacteria. Substantial clinical evidence is available to demonstrate protection against intestinal infections from S.aureus, Shigella, Salmonella, Vibrio cholerae, E. coli, rotavirus, Campylobacter, and protozoa. 81 Two facilitory actions of breast milk are apparent. The first encourages the growth of L. bifidus and thus crowds out the growth of other bacteria. In the second, the number of pathogens is also kept low by the direct action of lysozyme and lactoferrin. When the number of pathogenic bacteria is kept low, immune antibodies can keep the growth of potentially pathogenic bacteria under control and limit the invasion of bacteria through the gut wall into the bloodstream.

The intestinal flora of formula-fed infants is made up of predominantly gram-negative bacteria, especially coliform organisms, Bacteroides, and including Clostridium, Enterobacter, and Enterococcus. 282 Studies have demonstrated that potentially four distinctly different "microhabitats" for microflora within the GI lumen, within the mucus layer, separately within crypt mucus, and directly on the surface of the intestinal epithelium. The significance of the microhabitats and the effect of specific microorganisms has yet to be determined. 75 At weaning, the facultative anaerobes decline in number and obligate anaerobes (Bacteroides) become the predominant organisms in the intestine. Preterm infants are colonized with different types and numbers of bacteria from full term infants. The environment of neonatal intensive care units (NICUs), including incubators, widespread use of antibiotics and parenteral nutrition, and illness influence the microbial colonization. The short-and long-term effects of the different and changing GI microbiota are a concern. 186,206 This is particularly true when one considers the contributing factors or causes for sepsis, NEC, chronic lung disease, or poor neurologic outcome in NICUs.

The question of a causal role of intestinal microflora and the development of NEC in premature and very-low-birth-weight (VLBW) infants has been proposed. 123, 174 Gewolb et al 79 suggested that a low percentage of Bifidobacterium and Lactobacillus in the stool of VLBW infants within the first month of life is a risk factor for infection. Some studies on the use of probiotics and the occurrence of NEC have demonstrated a lower incidence of NEC in infants receiving probiotics. 19, 124, 162 Here again, the use of mothers' breast milk for premature infants and VLBW infants decreases the risk to the infant of sepsis, NEC, and infection-related events. 73, 245 In a controlled prospective study of high-risk, low-birth-weight infants in India using donor human milk, significantly fewer infections and no major infections were found in the group receiving human milk, although the control infants experienced diarrhea, pneumonia, septicemia, and meningitis. 185 

The properties of human milk do appear to control or limit infections in infants. Hundreds of articles 168, 185 have been written about the protective effect of breastfeeding, including the recent Agency for Healthcare Research and Quality publication on Breastfeeding and Maternal and Infant Health Outcomes in Developed Countries from 2007. 126 Using evidence-based analyses, the report documents the decreased risk for acute otitis media, GI infections, and lower respiratory tract diseases in breastfed infants in developed countries. 126 

Breast milk IgA has antitoxin activity against enterotoxins of E. coli and V. cholerae that may be significant in preventing infantile diarrhea. Antibodies against O antigen of some of the most common serotypes of E. coli were found in high titer in breast milk samples collected from healthy mothers in Sweden. The infants who had consumed reasonable amounts of breast milk with high titers of E. coli antibodies had antibodies in their stool. 106 Protection against cholera in breastfed children by antibodies in breast milk was studied by Glass et al. 81 A prospective study in Bangladesh showed cholera antibody levels to vary in the colostrum and milk.

The correlation among colonization, disease, and milk antibodies led the authors to conclude that breast milk antibodies against cholera do not protect children from colonization with V. cholerae, but they do protect against disease.

Salmonella infection was similarly studied by France et al 71 to evaluate the immunologic mechanisms in host colostrum and milk specific for salmonellae. Vigorous responses of colostral and milk cells against these organisms and nonspecific opsonizing capacity of the aqueous phase of colostrum and milk were demonstrated.

Gothefors et al 95 showed that E. coli isolated from stools of breastfed infants differed from strains found in formula-fed infants in two respects. First, E. coli strains were more sensitive to the bactericidal effect of human serum. Second, and more often, spontaneously agglutinated bacteria from other sites, such as the prepuce or periurethral area, were less sensitive in breastfed infants. These findings support the theory that breast milk favors proliferation of mutant strains, which have decreased virulence. This mutation of bacterial strains is another way breastfeeding may protect against infection.

It has been suggested that "milk immunization" is a dynamic process because a mother' s milk has been found to contain antibody to virtually all her infant' s strains of intestinal bacteria. The mother exposed to the infant' s microorganisms through either the breast or the gut responds immunologically to those microorganisms and thus directly provides protection for her immunologically immature infant.

The orderly review of data on the presence of antibodies in human milk has produced a substantial list of affected organisms. In addition to E. coli, antibodies to Bacteroides fragilis, Clostridium tetani, Haemophilus pertussis, Diplococcus pneumoniae, Corynebacterium diphtheriae, Salmonella, Shigella, Chlamydia trachomatis, V. cholerae, S. aureus, and several strains of Streptococcus (Table 5-9 ) have been identified. Noguera-Obenza and Cleary 196 have summarized the contribution of sIgA in breast milk in protecting infants from bacterial enteritis.

A study in Oslo by Hanson 106 of an outbreak of severe diarrhea caused by E. coli strain 0111 showed that six severely ill children were formula fed. Two infants who were breastfed had E. coli strain 0111 in their stools but showed few symptoms. Their mothers had no detectable antibodies for strain 0111 in their milk, which would suggest that other factors in human milk protect the infant from serious illness when no antibodies are in the milk. Hanson 106 also reported the results of another study in which after colonization with a specific strain of E. coli, mothers had large numbers of lymphoid cells in their milk with antibodies to that E. coli. The mothers' serum showed no such response. This supports the concept that antigen-triggered lymphoid cells from Peyer patches seek out lymphoid-rich tissue, producing IgA in the mammary gland. The mother is immunized in the gut at the same time as her milk. It has also been shown that E. coli enteritis can be cured by feeding human milk. Schlesinger and Covelli 246 studied possible cell-mediated immunity in breastfed infants. They showed that tuberculin-positive nursing mothers had reactive T cells in their colostrum and early milk. Furthermore, 8 of 13 infants nursed by tuberculin-positive mothers had tuberculinreactive peripheral blood T cells after 4 weeks. Cord blood had no such activity. No clinical or research data suggesting a protective effect of this apparently induced tuberculin reactivity in infants are available.

Protection against viruses has been the subject of similar studies. Breast milk contains antibodies against poliovirus, coxsackievirus, echovirus, enterovirus, influenza virus, reovirus, RSV, rotavirus, and rhinovirus. 187, 237 It has been confirmed that human milk inhibits the growth of these viruses in tissue culture. Nonspecific substances in human milk are active against arbovirus and murine leukemia virus, according to work by Fieldsteel. 67 A high degree of antiviral activity against Japanese B encephalitis virus as well as the two leukemia viruses has been found in human milk. The factor was found in the fat fraction and was not destroyed by extended heating, which distinguishes it from antibodies. May 168 believes the nonimmunoglobulin macromolecule antiviral activity in human milk is caused by specific fatty acids and monoglycerides (Table 5-10) . It is important to recognize other factors besides immmunoglobulins are contained in breast milk that can play a role in protection of the breastfeeding infant from viral infections. 102, 104 Specimens of human colostrum have been found to contain neutralizing activity against RSV. RSV has become a major threat in infancy and is the most common reason for hospitalization in infancy in some developed countries. It has a high mortality rate. Epidemics have occurred in special care nurseries. Statistically significant data collected by Downham et al 58 showed that, compared with uninfected control subjects who were breastfed (46 of 167), few breastfed babies (8 of 115) were among the infants hospitalized for RSV infection.

Fishaut et al 68 studied the immune response to RSV prospectively in 26 nursing mothers during several months. Antiviral IgM and IgG were rarely found in colostrum or milk. RSV-specific IgA, however, was identified in 40% to 75% of specimens.

Two mothers with the disease had specific IgG, IgM, and IgA antibody in serum and nasopharyngeal secretions, but only IgA was found in their milk. This confirms that IgA antibody to specific respiratory tract pathogens is present in the products of lactation. Because RSV appears to replicate only in the respiratory tract, the authors suggest that viral-specific antibody activity in the mammary gland may be derived from the bronchioleassociated lymphoid tissue.

In human milk, bile salt-stimulated lipase has been found to be the major factor inactivating protozoans (Table 5-11 ). 168 The mechanism by which lipase acts is not known, although it may generate fatty acids and monoglycerides, which inactivate enveloped bacteria, viruses, or protozoa. A nonimmunoglobulin, nonlipase, heat-stable factor has been identified in human milk that can inactivate Giardia lamblia.

Human milk protects against many intestinal and respiratory pathogens with minimal evidence of inflammation. Goldman et al 90 hypothesize that human milk is poor in initiators and mediators of inflammation but rich in antiinflammatory agents. Several major biochemical pathways of inflammation, including the coagulation system, the fibrinolytic system, and complement, are poorly represented in human milk. Box 5-4 outlines the antiinflammatory properties of various constituents and the paucity of certain proinflammatory mediators in breast milk. The interaction of factors in the milk with one another or with host defenses cannot be entirely predicted by examining each factor separately. When the decreased response of human milk leukocytes to chemoattractant peptides was demonstrated by Thorpe et al, 262 the failure of the response of human milk leukocytes was not caused by alterations in maternal peripheral blood leukocytes. This suggests that inhibitors are in the milk and that human milk leukocytes may be modified in the mammary gland to protect through noninflammatory mechanisms. 262 Only low numbers of basophils, mast dells, eosinophils, and cytotoxic T-cells are present in breast milk. Many other studies have documented the decreased function of milk polymorphonuclear leukocytes and macrophages in both colostrum and mature milk. 30 The antioxidant properties of human colostrum were demonstrated by Buescher and McIlheran 31 using aqueous human colostrum on human PMNs. The colostrum significantly interfered with PMN oxygen metabolic and enzymatic activities that are important in the mediation of acute inflammation.

Antioxidants in breast milk can also contribute to the overall antiinflammatory effect of breast milk. Demonstrated antioxidants contained in breast milk include an ascorbate-like compound, uric acid, α-tocopherol, β-carotene, and l-histidine, all of which scavenge oxygen radicals. Blood levels of α-tocopherol and β-carotene are higher in breastfed than unsupplemented formula-fed infants. Catalase, glutathione peroxidase, and lactoferrin are functionally antioxidants. Antioxidant activity has been demonstrated in colostrum and at lower levels in mature human milk.

Additionally, specific cytokines that can exhibit antiinflammatory effects have also been identified in human colostrum and milk: TGF-β 1 and -β 2 195,239,259 and IL-10. 74 A cytokine antagonist, IL-1RA, and soluble receptors for TNF-α are also found in colostrum and milk. 32, 254 Palkowetz et al 209 have reported that IL-1RA can decrease the action of IL-1β.

Both human colostrum and milk cause a diminished influx of polymorphonuclear cells to a local site of inflammation in two different in vivo models of inflammation in rats. 38, 96, 181 The inflammatory response can be protective for the host at the same time as it can produce the symptoms of clinical illness. Breast milk contains a large variety of antimicrobial factors that exert their protective effect without causing significant inflammation (e.g., sIgA, oligosaccharides, lactoferrin, nucleotides). Many other cells and factors in breast milk participate in a complex interaction to both protect the infant and limit the potential damaging effects of an uncontrolled inflammatory response. Further study into the dynamic interplay of the many factors in breast milk with developing infants' mucosal barriers and immune systems is needed to fully understand the protective immune response and the antiinflammatory benefits of human milk.

(See Chapter 17 on human milk as prophylaxis in allergy)

In discussing the allergic protective properties of human milk, it is difficult to identify specific protective factors that are proved to protect against allergy. It is equally difficult to discuss the proposed mechanisms of protection because the exact mechanism of "oral tolerance" remains theoretic and the relative importance of contributing factors to hypersensitivity must still be adequately defined. Some of the important variables concerning tolerance and sensitization are genetic background of the host, nature and dose of the antigen, frequency of exposure, timing (age) at first and subsequent exposures, immunologic status of the host, and route of exposure.

During the neonatal period the small intestine has increased permeability to macromolecules. Infants have more serum and secretory antibodies against dietary proteins than children or adults. Production of IgA in the intestinal tract is delayed until 6 weeks to 3 months of age. IgA in colostrum and breast milk prevents the absorption of foreign macromolecules when an infant' s immune system is immature. Mucin, oligosaccharides, and other factors within breast milk may affect antigen presentation. Protein of breast milk is species specific and therefore nonallergic for human infants.

No antibody response has been demonstrated to occur with human milk in infants. It has also been shown that macromolecules in breast milk are not absorbed.

Indirect evidence can be inferred from a demonstration of an infant' s response to cow milk protein. Within 18 days of taking cow milk, the infant will begin to develop antibodies. Since the advent of prepared formulas, in which the protein has been denatured by heating and drying, the incidence of cow milk allergy has been considered to be 1%. The most reliable means of diagnosing cow milk allergy is by challenging with isolated cow milk protein.

Although circulating antibodies and coproantibodies have been identified, these are not reliable techniques for a clinician involved in patient care.

The allergic syndromes that have been associated with cow milk allergy include gastroenteropathy, atopic dermatitis, allergic rhinitis, chronic pulmonary disease, asthma, eosinophilia, failure to thrive, and sudden infant death syndrome, or cot death, which has in some cases been attributed to anaphylaxis to cow milk. 132, 156 GI symptoms have received the greatest attention and include spittingup, colic, diarrhea, blood in the stools, frank vomiting, weight loss, malabsorption, colitis, and failure to thrive. Cow milk has been associated with GI protein and blood loss. The diagnosis is best made by elimination of cow milk from diet and, when appropriate, challenge tests. Cutaneous testing is of little help. Cow milk allergy has been described in breastfed infants, and exclusive breastfeeding alone is not sufficient to protect an infant at high risk to become sensitized to cow milk proteins. 129 The incidence of cow milk allergy in exclusively breastfed infants has been estimated as 0.4% to 0.5% compared with the overall incidence ranging from 1.9% to 7.5% in population-based studies. 129 Murray 182 showed the association of nasal secretion eosinophilia with infants freely fed cow milk or solid foods compared with eosinophilia in strictly breastfed infants. In infants receiving cow milk, 32% had high eosinophilic secretions, and only 11% of breastfed infants had eosinophils present in nasal secretions.

Not surprisingly, many different antigenic specificities are recognized when the colostrum or milk of one species is fed to or injected into another species. Cow milk is high on the list of food allergens, particularly in children. Sensitivity to cow milk is responsible for at least 20% of all pediatric allergic conditions, according to Gerrard. 78 Evidence indicates that IgA antibodies play an important role in confining food antigens to the gut. Food antigens given to bottle-fed infants before they can make their own IgA, and when they are deprived of that in human milk and the plasma cells, may be expected to be more readily absorbed. Glaser 80 first made the association between the drop in breastfeeding and the rise in allergy. He pioneered the theory of prophylactic management of allergy. Allergy in infancy is associated with a familial history of atopic disease and elevated cord blood IgE levels. The introduction of "foreign" proteins to an infant' s diet and even to the mother' s diet in the breastfeeding dyad can lead to allergic symptoms in the infant. Exclusive breastfeeding does not protect high-risk children from allergic symptoms unless the mother also adheres closely to a restrictive diet that excludes common allergens. 5, 12 A large body of literature examines whether breastfeeding protects against atopic disease. In 1988, Kramer 148 defined 12 standards for methodology and the study of allergy and breastfeeding. The standards clarified the definitions of breastfeeding, measurable outcomes, and the diagnostic criteria for specific allergic syndromes, defined children at high risk for atopic disease, and addressed methods to decrease bias and control for confounding variables. Several recent large meta-analyses have been performed assessing the protective effect of breastfeeding against allergic rhinitis, atopic dermatitis, and asthma. 76, 77, 175 Exclusive breastfeeding during the first 3 months of life protected against allergic rhinitis (summary odds ratio 0.74; 95% confidence interval 0.54 to 1.01) with or without a family history of atopy. 175 Exclusive breastfeeding for at least 3 months was associated with lower rates of atopic dermatitis in children with a family history of atopy. 76 Exclusive breastfeeding in the first months of life was protective against asthma during childhood (odds ratio 0.70; 95% confidence interval 0.60 to 0.81). 77 Chapter 17 discusses the prophylactic management of the potentially allergic infant.

The major elements in human milk related to the infant' s immune system are direct-acting antimicrobial factors, antiinflammatory factors, and immunomodulating bioactive compounds. 108 Epidemiologic studies have produced compelling information that suggests that breastfeeding for 4 months or longer can provide some immunologic protection against some childhood-onset diseases. 85, 91, 169 In 1991, Viirtanen et al 268 reported a prospective long-term study among children in Finland that showed a significantly lower incidence of type 1 diabetes in those at-risk children who had been breastfed for 4 months or longer. Other epidemiologic studies have demonstrated a decreased incidence of type 1 insulin-dependent diabetes mellitus in breastfed children. 20, 169, 197 These clinical observations have been supported in the laboratory by studies of diet control in diabetic mice. The isolation of a bovine albumin peptide as a possible trigger of type 1 insulin-dependent diabetes mellitus makes further study imperative. 136 Based on limited data, the recent Agency for Healthcare Research and Quality report cautiously concluded that breastfeeding for at least 3 months reduced the risk for type 1 diabetes compared with breastfeeding for less than 3 months. For type 2 diabetes, the same report concluded that breastfeeding in infancy produced a decreased risk compared with not breastfeeding. 126 The review of the national perinatal collaborative study by Davis et al 54 showed a protective effect against development of childhood cancer by being breastfed for 4 months or longer for children followed for 10 years. The effect was greater for acute leukemia and lymphoma. The role of infant feeding practices showed a similar effect of breastfeeding as protective in postponing or decreasing the occurrence of inflammatory bowel disease in childhood. 145, 232 Greco et al 97 reported a decreased risk for celiac disease in breastfed infants. The AHRQ report concluded that an association exists between breastfeeding for at least 6 months and a decreased risk for developing acute lymphocytic leukemia and acute myelogenous leukemia. 126 Maternal renal allografts have a better survival rate in individuals who were breastfed in infancy compared with those who were not breastfed. 37, 143 The mechanism of these apparent long-term immunologic benefits remains unclear, although theories abound. 85 Given the potential for confounding factors and bias in large long-term studies, confirmation of these proposed benefits by additional carefully controlled trials is required.

An increasing amount of accumulated epidemiologic literature, utilizing improved methodology and statistics, demonstrates the protective benefits of human milk for infants. A large number of bioactive factors have been identified and measured in breast milk during the period of lactation. Additional research is needed to clarify the interactions and the mechanisms of action of the many bioactive factors in human milk and then correlate these immunomodulatory actions with specific protective benefits for the infant.

Section on Breastfeeding: Breastfeeding and the use of human milk

Activity of the alternative pathway of complement in the newborn infant

A prospective cohort study on breast-feeding and otitis media in Swedish infants

Exclusive breastfeeding reduces acute respiratory infection and diarrhea deaths among infants in Dhaka Slums

Sensitization to common allergens and its association with allergic disorders at age 4 years: A whole population birth cohort study

Breast feeding and protection against neonatal sepsis in a high risk population

Breastfeeding and the capital risk of hospitilization for respiratory disease in infancy

Human chemokines: An update

Infant Feeding patterns and risks of death and hospitalization in the first half of infancy: multicentre cohort study

IgE and IgD in human colostrum and plasma

Diet and faecal flora in the newborn: Nucleotides

Influence of dietary manipulation on incidence of atopic disease in infants at risk

Studies of breastfeeding and infections. How good is the evidence?

Relation between infant feeding and infections during the first six months of life

Positional otitis media

Immunologic benefits and hazards of milk in maternal-perinatal relationship

Influence of breast-feeding on the bifid flora of the newborn intestine

Bactericidal activity of human milk leukocytes

Oral probiotics prevent necrotizing enterocolitis in very low birth weight neonates

Relation between breast-feeding and incidence rates of insulin-dependent diabetes mellitus: A hypothesis

Development and basic mechanisms of human gut immunity

Mucosal immunity: Integration between mother and the breast-fed infant

Mucosal and glandular distribution of immunoglobulin components: differential localization of free and bound secretory component in secretory epithelial cells

Ontogeny of the mucosal immune system and IgA deficiency

The B-cell system of human mucosae and exocrine glands

Milk derived GM 3 and GD 3 differentially inhibit dendritic cell maturation and effector fuctionalities

The epithelial cells and cell fragments in human milk

Effect of dietary nucleotide supplementation on diarrhoeal disease in infants

Interleukin-2 in human milk: A potential modulator of lymphocyte development in the breastfed infant

Bioactive Components of Human Milk

Antioxidant properties of human colostrum

Soluble receptors and cytokine antagonists in human milk

Polymorphonuclear leukocytes in human colostrum and milk

Developmental mucin gene expression in the gastroduodenal tract and accessory digestive glands. II. Duodenum and liver, gallbladder, and pancreas

Iron-binding proteins and other factors in milk responsible for resistance to Escherichia coli

Daily ingestion of immunologic components in human milk during the first four months of life

Breast feeding and maternal-donor renal allografts: Possibly the original donor-specific transfusion

Endotoxin and hypoxiainduced intestinal necrosis in rats: The role of platelet activating factor

Advances in nutritional modifications of infant formulas

Dietary nucleotides: cellular immune, intestinal and hepatic system effects

Dietary nucleotide effects upon immune function in infants

The role of nucleotides in human nutrition

Full breastfeeding duration and associated decrease in respiratory tract infection in US children

Bioactive Components of Human Milk

Decreased interleukin-10 production by neonatal monocytes and T cells: relationship to decreased production and expression of tumor necrosis factor-alpha and its receptors

Changes in carbohydrate composition in human milk over 4 months of lactation

Oligosaccharides from human milk block binding and activity of the Escherichia coli heat-stable enterotoxin (Sta) in T84 intestinal cells

Identification of Nestin-positive punitive mammary stem cells in human breastmilk

Regulation of cytokine release from mononuclear cells by the iron-binding protein lactoferrin

Breastfeeding reduces risk of respiratory illness in infants

Role of oligosaccharides and glycoconjugates in intestinal host defense

Persistence of human milk proteins in the breast-fed infant

Infant feeding and childhood cancer

Differences in morbidity between breast-fed and formula-fed infants

Partition of nitrogen intake and excretion in low-birth-weight infants

Maternal exposure to endocrine-active substances and breastfeeding

Breast-feeding protects against respiratory syncytial virus infections

The effects of infant feeding on rotavirus-induced gastroenteritis: A prospective study

Exclusive breastfeeding protects against bacterial colonization and day care exposure to otitis media

Exclusive breast-feeding for at least 4 months protects against otitis media

Delayed breastfeeding initiation increases risk of neonatal mortality

Deficient classical complement pathway activity in newborn sera

Variability of human milk neutral oligosaccharides in a diverse population

Effect of storage and heat on antimicrobial proteins in human milk

Intestinal microflora in early infancy: composition and development

Nonspecific antiviral substances in human milk active against arbovirus and murine leukemia virus

Bronchomammary axis in the immune response to respiratory syncytial virus

The role of gut-associated lymphoid tissues and mucosal defense

Influence of the heat treatment of human milk on some of its protective constituents

Breast-feeding and Salmonella infection

Iron in human milk

The effect of maternal milk on neonatal morbidity of very low-birth-weight infants

Expression of functional immunomodulatory and anti-inflammatory factors in human milk

Impact of the intestinal microbiota on the development of mucosal defense

Breastfeeding and the onset of atopic dermatitis in childhood: A systematic review and meta-analysis of prospective studies

Breast-feeding and the risk of bronchial asthma in childhood: A systematic review with meta-analysis of prospective studies

Sensitization to substances in breast milk: Recognition, management and significance

Stool microflora in extremely low birthweight infants

The dietary prophylaxis of allergic disease in infancy

Protection against cholera in breast-fed children by antibodies in breast milk

Antibodyforming cells in human colostrum after oral immunisation

Rapid hightemperature treatment of human milk

Immunologic system in human milk

The immune system of human milk: Antimicrobial, antiinflammatory and immunomodulating properties

Modulation of the gastrointestinal tract of infants by human milk: Interfaces and interactions-An evolutionary perspective

Evolution of immunologic functions of the mammary gland and the postnatal development of immunity

Spectrum of immunomodulating agents in human milk

Immunologic components in human milk during the second year of lactation

Anti-inflammatory properties of human milk

Cytokines in human milk: properties and potential effects upon the mammary gland and the neonate

Immunologic factors in human milk during the first year of lactation

Immunologic protection of the premature newborn by human milk

Lactoferin structure, function and applications

Breast feeding and biological properties of faecal E. coli strains

Antiinflammatory effects of human milk on chemically induced colitis in rats

Case control study on nutritional risks in celiac disease

Free secretory components and lactoferrin of human milk inhibit the adhesion of enterotoxigenic Escherichia coli

Possible influence of vitamin B 12 -binding protein in milk on the intestinal flora in breast-fed infants. II. Contents of unsaturated B 12 -binding protein in meconium and faeces from breast-fed and bottle-fed infants

A hitherto unrecognized biochemical difference between human milk and cow' s milk

Mother' s milk and sewage: Their interactive effects on infant mortality

Antiviral activity of purified human breastmilk mucin

Bioactive factors in human milk

Protective function of human milk: the milk fat globule

Immunoglobulins in feces from infants fed human or bovine milk

The mammary gland as an immunological organ

Host defense of the neonate and the intestinal flora

The state of health of children in the developing world

The role of breastfeeding in prevention of neonatal infection

Immune system modulation by human milk

Breastfeeding influences thymic size in late infancy

Effect of long term consumption of probiotic milk on infections in children attending day care centres: Double blind, randomised trial

Cytokines (IL-1[beta], IL-6, TNF-[alpha], TGF-[beta]1, and TGF-[beta]2) and prostaglandin E2 in human milk during the first three months postpartum

B cell function in the newborn

The Immunologic Role of Viable Leukocytic Cells in Mammary Gland/Human Lactation/Milk Synthesis

Host defense benefits of breastfeeding for the infant

Assays for endogenous components of human milk: Comparison of fresh and frozen samples corresponding analytes in serum

Human colostral cells: Phagocytosis and killing of E. coli and C. albicans

Angiogenins: a new class of microbicidal proteins in in innate immunity

Postnatal maturation of immune competence during infancy and childhood

Xenobiotics and breastfeeding

Protective effect of breast feeding against infection

Quantitative changes in faecal microflora preceding necrotizing enterocolitis in premature neonates

Reduced incidence of necrotizing enterocolitis associated with enteral administration of Lactobacillus acidophilus and Bifidobacterium infantis to neonates in an intensive care unit

Breast milk macrophages spontaneously produce granulocyte-macrophage colony-stimulating factor and differentiate into dendritic cells in the presence of exogenous interleukin-4 alone

Breastfeeding, Maternal and Infant Health Outcomes in Developed Countries

Viral, nutritional, and bacterial safety of flash-heated and pertoria-pasteurized breast milk to prevent mother-tochild transmission of HIV in resource-poor countries

Bacterial safety of flash-heated and unheated expressed breastmilk during storage

Development of cow' s milk allergy in breast-fed infants

Immunological protection of the neonatal gastrointestinal tract: The importance of breast feeding

The effect of Pretoria pasteurization on bacterial contamination of hand-expressed human breast milk

Dietary prophylaxis of allergic disease in children

How many child deaths can we prevent this year?

Human milk stimulates B cell function

Enhancement of the circulating antibody secreting cell response in human diarrhea by a human lactobaccillus strain

A bovine albumin peptide as a possible trigger of insulin-dependent diabetes mellitus

Activated neutrophils and neutrophil activators in human milk: Increased expression of CD11b and decreased expression of L-selectin

T cell subsets in human colostrum

IgD in human colostrum

Commensal gut bacteria: mechanisms of immune modulation

Effects of nutritional supplements on anti-infective factors in human milk

The enteromammary immune system: An important new concept in breast milk host defense

Influence of breast feeding on subsequent reactivity to a related renal allograft

Growth Factors and Cytokines in Milk

Role of infant feeding practices in development of Crohn' s disease in childhood

Fecal secretory immunoglobulin A in breast milk versus formula feeding in early infancy

Infant feeding, infection, and public health

Cytokine profiling human colustrum and milk by protein array

Toward consistency in breastfeeding definitions

Innate recognition of bacteria in human milk is mediated by a milk-derived highly expressed pattern recognition receptor, soluble CD14

Kolsto Otnaess AB: Trace amounts of ganglioside GM1 in human milk inhibit enterotoxins from Vibrio cholerae and Escherichia coli

Efecto de la suplementacion dietetica con nucleotidos sobre la diarrea en el lactante sano

Human breast milk: current concepts in immunology and infectious diseases

Total potentially available nucleosides of human milk by stage of lactation

Soluble forms of Toll-like receptor (TLR)2 capable of modulating TLR2 signaling are present in human plasma and breast milk

Cow' s milk protein allergy

Lactoferrin structure and functions

Impaired innate immunity in the newborn: newborn neutrophils are deficient in bactericidal/permeability-increasing protein

The influence of parity, age and maturity of pregnancy on antimicrobial proteins in human milk

Alterations of lymphocytes and of antibody content of human milk after processing

Review: Nomenclature and biologic significance of cytokines involved in inflammation and the host immune response

Oral probiotics reduce the incidence and severity of necrotizing enterocolitis in very low birth weight infants

Lactoferrin in milk

Nutritional roles of lactoferrin

Effect of exercise on immunologic factors in breast milk

Chemokines-Chemotactic cytokines that mediate inflammation

Lower respiratory illness in infants and low socioeconomic status

Antimicrobial properties and microbial contaminants of breast milk: An update

Reduced risk of IDDM among breast-fed children. The Colorado IDDM Registry

An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system

Antibodies in milk

Leukocyte function and the development of opsonic and complement activity in the neonate

The antimicrobial activity of human colostral antibody in the newborn

Enterobacteriaceae and neonatal necrotizing enterocolitis

Gdalevich M: Does breastfeeding protect against allergic rhinitis during childhood? A meta-analysis of prospective studies

Effect of maternal nutritional status on immunological substances in human colostrum and milk

IgG subclasses: development of the serum concentrations in "normal" infants and children

Intestinal microflora of human infants and current trends for its nutritional modulation

Breastfeeding patterns, time to initiation, and mortality risk among newborns in Southern Nepal

Innate defenses of the intestinal epithelial barrier

Human colostrum has antiinflammatory activity in a rat subcutaneous air pouch model of inflammation

Infant feeding and respiratory allergy

Role of epidermal growth factor and other growth factors in the prevention of necrotizing enterocolitis

Influence of dietary nucleotides on plasma immunoglobulin levels and lymphocyte subsets of preterm infants

The value of human milk in the prevention of infection in the high-risk low-birth-weight infant

Microbes and the developing gastrointestinal tract nutrition in clinical practice

Antiviral components of human milk

Human glycoconjugates that inhibit pathogens

Neonatal protection by an innate immune system of human milk consisting of oligosaccharides and glycans

Oligosaccharides and glycoconjugates in human milk: Their role in host defense

Neutral glycolipids of human and bovine milk

Carbohydrates in Milks: Analysis, Quantities and Significance

Human milk glycans protect infant against enteric pathogens

Improved thymic function in exclusively breastfed infants is associated with higher interleukin 7 concentrations in their mothers breastmilk

Transforming growth factor activity in human colostrum

FW: A meta-analysis of infant diet and insulin-dependent diabetes mellitus: Do biases play a role

Structure and biological actions of lactoferrin

breast feeding and respiratory morbidity in infancy: A birth cohort study

Role of transforming growth factor-β in breast milk for initiation of IgA production in newborn infants

Immunologic aspects of human colostrum and milk. II. Characteristics of lymphocyte reactivity and distribution of E-rosette forming cells at different times after the onset of lactation

Immunologic aspects of human colostrum and milk. I. Distribution characteristics and concentrations of immunoglobulins at different times after the onset of lactation

Immunologic aspects of human colostrum and milk: interaction with the intestinal immunity of the neonate

Techniques for the storage of human breastmilk: Implications for anti-microbial functions and safety of stored food

Alterations in intestinal microbial flora and human disease

Inhibition of enterotoxin from Escherichia coli and Vibrio cholerae by gangliosides from human milk

Relation of infant feeding practices, cigarette smoke exposure, and group child care to the onset and duration of otitis media with effusion in the first two years of life

Production of interleukin-6 and interleukin-8 by human mammary gland epithelial cells

Development of the human infant intestinal microbiota

Development of the gastrointestinal mucosal barrier. Evidence for structural differences in microvillus membranes from newborn and adult rabbits

In vitro studies on the T-lymphocyte population of human milk

Quantitative determination of immunoglobulins, lysozyme, and certain electrolytes in breast milk during the entire period of lactation, during a 24-hour period, and in milk from the individual mammary gland

Probiotic bacteria down-regulate the milk-induced inflammatory response in milk-hypersensitive subjects but have an immunostimulatory effect in healthy subjects

Factors influencing the composition of the intestinal microbiota in early infancy

Bacterial imprinting of the neonatal immune System: lessons from maternal cells?

Control of lymphocyte recirculation in man. II. Differential regulation of the cutaneous lymphocyte-associated antigen, a tissue-selective homing receptor for skin-homing T cells

Modulation of the immune system by human milk and infant formula containing nucleotides

Human milk humoral immunity and infant defense mechanisms

Breast-feeding and urinary tract infection

The milk mononuclear phagocyte

Tight junctions in epithelial cells of human fetal hindgut, normal colon, and colon adenocarcinoma

Breast-feeding and diarrheal morbidity

Do infants need nucleotide supplemented formula for optimal nutrition?

Effects of microwave radiation on anti-infective factors in human Milk

Breastfeeding and hospitilization for diarrheal and respiratory infection in the United Kingdom Millenium Cohort Study

Breast-feeding and infant illness: A dose-response relationship?

Innate recognition of intracellular pathogens: Detection and activation of the first line of defense

Antimicrobial factors in human milk

Current concepts of infections of the fetus and newborn infant

Lymphocyte subsets in colostrum

Breast-feeding and maternal smoking in the etiology of Crohn' s disease and ulcerative colitis in childhood

Lactoferrin acts as an alarmin to promote the recruitment and activation of APCs and antigen-specific immune responses

Interleukin-6 in human milk

Cholera-like enterotoxin produced by Campylobacter jejuni

Breastmilkderived antigen-specific CD 8+ T cells: An extralymphoid effector memory cell population in humans

Maternal antibodies in breast milk protect the child from enterovirus infections

MUC1 In human milk blocks transmission of human immunodeficiency virus from dendritic cells to T cells

Transforming growth factor-beta (TGF-beta) in human milk

Paneth cells, defensins, and the commensal microbiota: a hypothesis on intimate interplay at the intestinal mucosa

Biological role of lactoferrin

Prolonged exclusive breast-feeding results in low serum concentrations of immunoglobulin G, A and M

A longitudinal analysis of infant morbidity and the extent of breastfeeding in the United States

Ribonucleotides: conditionally essential nutrients shown to enhance immune function and reduce diarrheal disease in infants

Randomized trial of donor human milk vs. preterm formula as substitutes for mothers' own milk in the feeding of extremely premature infants

Evidence for transmission of lymphocyte responses to tuberculin by breast-feeding

Inhibition of adhesion of S-fimbriated Escherichia coli to buccal epithelial cells by human milk fat globule membrane components: A novel aspect of the protective function of mucins in the nonimmunoglobulin fraction

Control of isotype switching by T cells and cytokines

Continuous culture selection of bifidobacteria and lactobacilli from human faecal samples using fructooligosacchirade as selective substrate

Functional analysis of neutrophil granulocytes from healthy, infected, and stressed neonates

Complement system in healthy term newborns: reference values in umbilical cord blood

Lactoferrin, a major defense protein of innate immunity, is a novel maturation factor for human dendritic cells

Delineation of the functional capacity of human neonatal lymphocytes

Cytokines in human milk

Diversity of the human gastrointestinal tract microbiota revisited

Secretory IgA against enterotoxins in breast-milk

Immunoglobulin secretion by the normal and the infected newborn infant

Activation and activity of the superoxide-generating system of neutrophils from human infants

Reciprocal IFN-gamma and TGF-beta responses regulate the occurrence of mucosal inflammation

Interleukin-18 in human milk

Antibodies in milk

Decreased response of human milk leukocytes to chemoattractant peptides

Antimicrobial peptides of lactoferrin

Determination of total potentially available nucleosides in human milk from Asian women

Human milk-derived B cells: A highly activated switched memory cell population primed to secrete antibodies

Dietary nucleotides, a requirement for helper/inducer T lymphocytes

CD14: A soluble pattern recognition receptor in milk

Infant feeding in Finnish children less than 7 yr of age with newly diagnosed IDDM. Childhood Diabetes in Finland Study Group

Early and late effects of breastfeeding: Does breast-feeding really matter?

Evolution of the mammary gland from the innate immune system

Immune components of colustrum and milk: A historical perspective

Ig isotype switching in B lymphocytes. The effect of T cellderived interleukins, cytokines, cholera toxin, and antigen on isotype switch frequency of a cloned B cell lymphoma

Basis and implications of selectively diminished cytokine production in neonatal susceptibility to infection

CCL28 Controls immunologlobulin (Ig)A plasma cell accumulation in the lactating mammary gland and IgA antibody transfer to the neonate

Does breast milk protect against septicaemia in the newborn?

Activated and memory T lymphocytes in human milk

The development of the complement system after 28 weeks' gestation

Breast feeding and lower respiratory tract illness in the first year of life. Group Health Medical Associates

The development of the immune system

Development of natural killer cytotoxicity during childhood: marked increases in number of natural killer cells with adequate cytotoxic abilities during infancy to early childhood

An increase in polymorphonuclear leucocyte chemotaxis accompanied by a change in the membrane fluidity with age during childhood

Development and differences of intestinal flora in the neonatal period in breastfed and bottle-fed infants

Scientific rationale and benefits of nucleotide supplementation of infant formula