Effects of Milk Secretory Immunoglobulin a on the Commensal Microbiota

Protective Factors in Human Milk

Published online: March 30, 2020 Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335)

Effects of Milk Secretory Immunoglobulin A on the Commensal Microbiota

a, b a b Vanessa P. Dunne-Castagna David A. Mills Bo Lönnerdal a Department of Food Science and Technology, University of California, Davis, CA, USA; b Department of Nutrition, University of California, Davis, CA, USA

Abstract Secretory immunoglobulin A (SIgA) is intimately involved in the transfer of maternal immunity to the newborn breastfed infant. Recent research demonstrates the significance of SIgA in the initial development of the newborn’s microbiota and in the establishment of a tolero- genic immunologic disposition towards nonpathogenic organisms and environmental antigens. SIgA has long been known to prevent pathogen binding to the host epithelium through immune exclusion involving numerous mechanisms. This process primarily involves T-cell-dependent, somatically hypermutated monoclonal antibodies with high specificity towards pathogen surface antigens, and the success of the immune response is dependent upon the specific antigen recognition. Whereas this role is important, there is an alternate, dual role for SIgA in the health of the host – protection and promotion of commensal colonization and maintenance of homeostatic immunity. This latter role is primarily dependent upon N- and O-glycan moieties lining the secretory component and heavy chain of the SIgA dimer, with interactions independent of immunoglobulin specificity. These SIgA molecules are nonspecific polyclonal antibodies generated from plasma cells activated by dendritic cell sampling of luminal contents in the absence of inflammation. Breast milk is the primary sup- ply of such polyclonal polyreactive SIgA in the initial stages of neonatal colonization, and it provides vital pathogen resistance while promoting colonization of commensal microbiota. © 2020 Nestlé Nutrition Institute, Switzerland/S. Karger AG, Basel Secretory IgA Production

In the mucosa, the majority of the immunoglobulin repertoire consists of somatically hypermutated IgA proteins, and over 25% are polyreactive to microbial-associated molecular patterns and other common antigens [1]. In addition, mucosal IgA production maintains a very high constant level in the absence of infection, up to 5 g/kg/day in an adult intestine [2], but it can be increased in an inflammatory environment. Recent studies using photoconversion to track Pey- er’s patch B cells showed marked small intestinal Peyer’s patch germinal center clonal exchange of memory B cells during a 3-day period [3], indicating consid- erable trafficking of memory B cells throughout mucosal surfaces in the gastrointestinal tract. In the mucosa, there seems to be a high level of low-affinity, di- versified germinal-center B-cell responses to microbial antigens that are required to maintain intestinal homeostasis. In addition to these T-cell-dependent responses which account for about 75% of IgA induction in the mucosa, B cells can be activated via T-independent processes. These may involve pattern recognition receptor signaling, as most IgA generated in mice deficient in T cells or with deficiencies in somatic hypermutation but without class switching produce primarily microbial-pattern (low- affinity) antibodies [4]. Interestingly, in these studies, the microbiota expanded into a dysbiotic state, with higher levels of Proteobacteria, suggesting the need for T-cell-dependent IgA responses to maintain intestinal homeostasis. Mature B cells are released into the circulation as plasmablasts and home back to the site of mucosal induction to secrete dimeric IgA. SIgA is derived upon transcytosis of dimeric IgA through mucosal epithelial cells bound to the polymeric immunoglobulin receptor (pIgR) on the basolateral membrane, and the subsequent apical cleavage of the immunoglobulin-bound secretory component (SC), which releases free SIgA into the lumen.

Immune Exclusion

Once in the lumen, there are a number of mechanisms by which SIgA functions and more are being discovered. Traditionally, it was thought that SIgA acts through agglutination and neutralization, excluding pathogens from binding to epithelial cells via Fab-dependent, high-affinity interactions to pathogen surface antigens. Although neutralization is supported by imaging of viral pathogens [5], agglutination has come into question recently as the concentration of an in- fectious agent is likely too low to provide enough localized pathogen to create immune complexes with SIgA. A recent study proposed a concept of “enchained

Milk SIgA and Commensals 159

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) growth” whereby bacterial pathogens coated with SIgA exhibit incomplete bi- nary fission and remain attached after replication [6]. In fact, this creates similar immune complexes as agglutination, preventing bacterial adhesion to epithelial cells and reducing pathology. In addition to these mechanisms, a novel function for Fab-dependent SIgA has been identified in work with the recombinant monoclonal SIgA, Sal4, directed at the acetylated O5 antigen of Salmonella enterica serovar Typhimurium (ST) [7]. In these studies, researchers found that Sal4 binding altered bacterial membrane integrity, compromising the cell en- ergy gradient. As proton motif force is needed to power the bacterial flagella, motility was inhibited, as was ATP production and overall metabolic processes. While it is still uncertain how exactly the binding of SIgA to LPS alters the membrane potential, it likely involves cross-linking of several O-antigens, as the use of a recombinant immunoglobulin single chain did not create this effect.

Glycan-Mediated Binding

Although the Fab-mediated interactions of SIgA are vital for reducing pathogenic infections in the mucosa, as mice deficient in class-switch recombination and somatic hypermutation (AID–/–) [4] who are unable to create high-affinity immunoglobulins demonstrate heightened susceptibility to infection, antigen- independent, glycan-mediated interactions are also important in the function of this molecule. SIgA is highly decorated with N-linked glycans: 2 sites on each heavy chain, 8 in total, plus 7 on the SC, 1 on the J-chain, and an additional 2 on the hinge region on IgA2 (Fig. 1a) [8]. Many studies have now demonstrated that these glycans play a pivotal role in pathogen clearance, antigen presentation via M cells, and commensal homeostasis. For example, enteropathogenic Escherich- ia coli binds terminal sialic acid residues on the glycoprotein with intimin adhesion proteins [9], and Clostridium difficile toxin A neutralization is mediated by glycans [10]. Rochereau et al. [11] showed that sialic acid residues on IgA2 N- glycans (Fig. 1b) are required to bind on the apical surface of M cells via dectin-1 for reverse transcytosis and subsequent antigen presentation in the subepithelial dome. In addition, subepithelial dome dendritic cells require SIgA glycans for binding these immune complexes via DC-SIGN, a C-type lectin receptor [12]. Not only are these glycans on SIgA necessary for mucosal response to pathogens, more recent studies have also shown the utility of these glycans in binding to commensal bacteria, promoting immune tolerance, and maintaining homeostatic regulation. The first evidence of glycan-mediated commensal binding was shown utilizing a SIgA specific for Shigella flexneri LPS and 3 commensal bacte-

160 Dunne-Castagna/Mills/Lönnerdal

F(ab’)2

Sialic acid Mannose Fucose Galactose Glycan isoforms: Antigen binding cleft: N-acetylglucosamine Pathogen binding – High-affinity pathogen immune exclusion binding – immune b exclusion Commensal binding – protection, colonization Low-affinity commensal binding – colonization? Mucin localization Homeostatic regulation with self-antigen binding c

Fig. 1. Schematic representing the SIgA molecule (a), an example of the glycan isoforms (b), and some of the functions of the glycans and the antigen-binding region (c) on the molecule.

ria: Bifidobacterium lactis, Lactobacillus rhamnosus, and E. coli Nissle 1917 [13]. Using fluorescence microscopy, researchers visualized associations of this nonspecific SIgA to all bacteria, but when pretreated with glycosyl hydrolase (N- glycosidase F) to remove the N-linked glycans, SIgA was not able to bind to the 2 gram-positive bacteria, though it remained bound to the gram-negative one. Although further studies are needed to confirm the glycan dependency of commensals to SIgA, research from our lab and others are supporting the antigen- independent association of SIgA with commensal bacteria. Nakajima et al. [14] validated nonspecific SIgA binding using OTII mice. In their study, ovalbumin (OVA)-specific T cells were transferred into T-cell-deficient (CD3–/–) mice who were then provided with OVA antigen daily to induce an OVA-specific humor- al response. Flow cytometry of fecal bacteria revealed complexes of commensal- SIgA-OVA, compared to commensal-SIgA complexes in wild-type counter- parts, which was recapitulated in vitro with hybridoma-derived OVA-specific IgA but not IgG. Our own studies utilizing flow cytometry and fluorescence microscopy have shown a concentration-dependent interaction of both hetero-

Milk SIgA and Commensals 161

1,000 1,000 1,000 1,000 0.017 0.016 0.12 10.7 0.12 35.1 0.30 91.7

800 800 800 800

600 600 600 600

400 400 400 400 FL2-W: Alexa 647 FL2-W: Alexa 647 FL2-W: Alexa 647 FL2-W: Alexa 647 200 200 200 200 SIgA (A650) SIgA 0.51 99.3 0.94 88.3 0.38 64.4 0.58 7.42 0 0 0 0 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 100 101 102 103 104 FL1-H: Syto 9 FL1-H: Syto 9 FL1-H: Syto 9 FL1-H: Syto 9

Syto9 (cells)

Fig. 2. Flow-cytometric scatter plot showing increased pooled milk SIgA associated with a Bifidobacterium species with increased concentration of SIgA (percent circled). Syto9 is a DNA chelator indicating live bacteria, and A650 is anti-human-IgA with Alexa 647 fluoro- phore. Concentration is per 107 CFU bacteria.

geneously pooled milk SIgA (Fig. 2) and recombinant monoclonal SIgA with select commensal bacteria and enteropathogens [unpubl. data]. The utility of the nonspecific SIgA association with commensals is still unclear, though these studies and others have given some indication. Work using Caco-2 colonocytes exposed to the commensal bacteria Lactobacillus and Bifidobacterium demonstrated a 3.4-fold increase in the binding of bacteria to these cells, a reduction in NF-κB-induced proinflammatory cytokine production, and induction of the tight junction binding protein occludin when the bacteria were first associated with SIgA [15]. Our research group has recapitulated these studies and also found a significant reduction in the neutrophil recruiting IL-8 cytokine when SIgA is first associated to the commensals prior to binding in vitro on colonocytes. In addition, our studies have found that nonspecific SIgA association with commensal bacteria improves colonization in BALBc mice when the complex is provided orally [unpubl. data]. Gnotobiotic mice deficient in pIgR (pIgR–/–) are unable to produce SC-coated IgA, but serum and luminal IgA levels remain unchanged. These mice demonstrated a heightened translocation of commensal bacteria to the mesenteric lymph nodes, and an induction of a systemic immune response when introduced to commensal intestinal flora from specific pathogen-free mice. The systemic IgG response to colonization with specific pathogen-free microbio-

162 Dunne-Castagna/Mills/Lönnerdal

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) ta was not seen in wild-type counterpart mice, demonstrating functional immune compartmentalization in the presence of SIgA [16]. Murine studies show that SIgA coating of commensal bacteria promotes colonization in the outer mucosal layer of the colon [17], preventing the stimulation of intraepithelial lymphocytes and epithelial pattern recognition receptors, and thereby reducing the induction of innate inflammatory responses to nonpathogenic bacteria. In the OTII-CD3–/– mouse work, researchers found that Fab-independent association of SIgA with Bacteroides thetaiotamicron altered the expression of several genes involved in carbohydrate utilization and in exopolysaccharide production [14]. This was also seen in a study by Donaldson et al. [18] investigating gene expression changes with Bacteroides fragilis upon nonspecific SIgA association, where they found a change in exopolysaccharide production that led to colonization of a different ecological niche in the small intestine. Our research group has shown a significant reduction in enteropathogenic invasion both in vitro in human colonocytes and in vivo in BALBc mice when SIgA-associated commensals are introduced into the cells or animal prior to pathogen challenge, but not when either the commensal or SIgA alone are provided [unpubl. data]. Corthesy and Phalipon [19] demonstrated that in the respiratory tract, tissue localization was dependent on the glycosylation of the SIgA molecule. Upon in- tranasal challenge with S. flexneri, the glycosylated SIgA molecule was viewed in close association with the mucosa, and dissemination of the pathogen was con- fined to the nasal cavity, but deglycosylation of SIgA led to pathogenic colonization in the deep lung alveoli. Gibbins et al. [20] demonstrated an association of SIgA with salivary mucin proteins that then bind with other epithelial cell-pro- duced mucin proteins, indicating a mucin-mucin-driven SIgA interaction. While research supports the association of glycosylated SIgA with the outer mucosal layer in the large intestine, Rogier et al. [17] showed through fluorescent microscopy that the mucin MUC2 protein in the innermost mucosal lining of the gut epithelium is important not in binding to SIgA, but in excluding this complex from interacting with the epithelia. The exact mechanisms of SIgA interacting with the loose outer mucosa remain unclear, but examinations of other mucosal sites within mammals demonstrate the importance of the carbohydrate moieties in maintaining associations with both microorganisms and the mucosa.

Breast Milk Secretory IgA

The gut of the neonate, upon parturition, contains a number of innate immune barriers, an underdeveloped gut-associated lymphoid tissue, and naïve adaptive immunity, which can take up to 10 days to become stimulated [21]. These con-

Milk SIgA and Commensals 163

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) ditions can lead to an imbalanced inflammatory response without proper guid- ance from the mother’s passive immune defenses in breast milk that aid in es- tablishing a regulated environment during the initial onslaught of microbial colonization. SIgA is the primary mucosal antibody found in the highest abundance over other immunoglobulins in milk, with concentrations up to 15 mg/ mL in colostrum and ∼1 mg/ml in mature milk [22], providing the breastfed infant 0.5–1 g/day. The origin of milk antibodies has been evaluated through many studies. Although antibody-secreting plasma cells can be detected in ex- creted milk [23], the production of SIgA and SIgM likely arises from secreting plasma cells in the basolateral region of the mammary gland to allow for transcytosis through the epithelium and release of SC-bound immunoglobulins. The induction site for these mammary-homing lymphocytes is predominantly the gut-associated lymphoid tissue, with migration through the established entero- mammary pathway [24]; early work in this field showed that radioactively la- beled IgA-secreting mesenteric lymphocytes were found in the mammary gland of lactating murine dams 3–10 times higher in number than lymphocytes of peripheral lymph node origin [25]. In addition, antigen-specific SIgA can be detected in milk following oral introduction of the antigen in lactating mothers [26]. Antibodies secreted in breast milk have been shown experimentally in ro- dents and epidemiologically in humans to prevent pathogenic infection in the infant gastrointestinal tract through targeted pathogen-specific SIgA both in infancy and extended through childhood [26]. Harris et al. [27] showed that C57BL/6 mouse pups were protected against the intestinal parasite Heligmoso- moides polygyrus when fostered on milk rich in pathogen-specific immunoglobulins (IgA and IgG), but not when the milk contained no H. polygyrus-specific immunoglobulins. Studies of breastfed human infants infected with the intestinal pathogen Campylobacter jejuni showed that milk containing antiflagellin or anti-outer membrane IgA antibodies [28] is protective against symptomatic C. jejuni infections. These studies and others have led to the exploration of a ther- apeutic SIgA approach: vaccinating mothers to promote protective SIgA antibodies in breast milk [29], a strategy warranting further investigation.

Milk Secretory IgA and Infant Commensal Colonization

Whereas the protective role of pathogen-targeted secretory milk antibodies is well established, recent studies have emerged focusing on the functional conse- quences of milk SIgA in association with commensals in the development of the newborn microbiota. Analyses of the milk antibody repertoire show significant

164 Dunne-Castagna/Mills/Lönnerdal

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) amounts of polyclonal autoreactive antibodies with low levels of somatic hypermutation evident [30]. Multiple studies have observed cross-reactivity of both antigen-targeted monoclonal antibodies and polyreactive autoantibodies to commensal microbiota of the small intestine [1, 14, 18]. Thus, there is a high likelihood of the ability for the diverse polyclonal antibody repertoire of breast milk to bind with low affinity to bacteria found both within the mammary gland and in the gut of the breastfed infant. Whereas to date methodology has not been developed to differentiate the maternally versus endogenously derived SIgA in the SIgA-coated fraction of the infant gut microbiota, studies utilizing BugFACS reveal distinct microbiota coated with SIgA under homeostatic conditions [31] or in malnourished Malawian infants [32]. Consistent with IgA-seq studies in adults [33], malnourished infants had a higher relative abundance of Enterobac- teriaceae associated with intestinal inflammation, which represented a significant portion of the SIgA-associated taxa. Conversely, healthy infants showed consistently high SIgA association in members of the genera Akkermansia and Clostridium, regardless of dietary intake or age (1–24 months). Likely, these observed variations in IgA-coated fractions in the infant gut are a reflection of the differences in IgA repertoire development under varying disease conditions in both the infant mucosa and also in the breast milk environment. A recent study investigated the impact of breast milk SIgA on the development of necrotizing enterocolitis (NEC), a disease with complex etiology and high mortality afflicting preterm infants born under 33 weeks of gestation [34]. The study found that NEC development was correlated with a loss of IgA association to Enterobacteriaceae – through unclear mechanisms. The significance of IgA was confirmed in a NEC mouse model where suckling mouse pups were fostered on dams deficient in total immunoglobulins (Rag–/–) or IgA (Igha–/–). These mice developed NEC at the same rate as formula-fed pups, whereas con- trol pups with milk sufficient in IgA had low rates of NEC development. Al- though milk has a plethora of immune components and oligosaccharides able to modify the microbiota and contribute to fortifying the immune system of the young infant, this study supports a crucial role for IgA in the preterm infant. Rogier et al. [35] sought to understand the difference in the function of maternally derived SIgA versus actively secreted endogenous IgA in the newborn gut. In their murine study, they found that the microbiome of adult mice, whose nursing mothers were deficient in pIgR and, therefore, provided no maternal SIgA during the suckling period (although milk IgA levels were unchanged), differed in key taxa from those who were exposed to SIgA during suckling. Specifically, they found a more pronounced presence of Proteobac- teria of the family Pasteurellaceae, and Firmicutes of the family Lachnospira- ceae. Analysis of the RNA component of intestinal epithelial cells of the same

Milk SIgA and Commensals 165

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) two mouse groups after challenge with epithelial-disrupting dextran sulfate sodium (DSS) showed two major groups of genes varying in expression level between the mice. Notably, many genes involved in cell repair were seen to be linked to DSS exposure and were up-regulated in adult mice with suckling exposure to SIgA, and a cluster of genes important in cell metabolism and growth that were not associated with DSS exposure were also up-regulated in the same group of mice. These reports, although based on a murine model, indicate the importance of early exposure to maternally derived SIgA to the long-term health of the child. Further research is needed to elucidate the mechanisms behind these results. In addition to these functions, SC may be important for protection against proteolytic degradation of the IgA antibody itself and perhaps the bacteria they coat, as evidenced by the presence of intact SIgA in the stool of exclusively breastfed infants [36] and in the 2011 study by Mathias and Corthésy [13], who showed that SC-bound, but not unbound, IgA resisted degradation in vitro. This protein-coating model may be one mechanism by which breast milk commensals survive intestinal digestion in order to colonize the gut of the infant. Indeed, our research group has found increased viability of commensal bacterial following in vitro intestinal digestion when the bacteria are first associated to pooled milk SIgA [unpubl. data].

Conclusion

SIgA plays important roles in the developing infant gut, both in pathogen de- fense and commensal homeostasis. Although breast milk provides a plethora of biologically active immune components and microbial-modifying factors, studies involving a reductionist approach to investigate individual bioactive molecules have helped elucidate the importance of the heavily glycosylated SIgA both in infancy and into adulthood. Studies are needed to determine the functional consequence of a SIgA-commensal complex both within the milk milieu and in the neonate gut with regard to commensal colonization and regulation of the infant mucosal immune response. Continued investigation into this relationship may translate into improvements in infant health and nutrition.

Acknowledgments

The authors wish to thank members of the Mills lab who contributed to the original research described in this study: Jassim Al-Oboudi, Eric Nonnecke, Hiu Ling Mak, Austin Ghera, Sarah Goldberg, and Meg Tseng.

166 Dunne-Castagna/Mills/Lönnerdal

Research was supported by a grant from the Bill and Melinda Gates Foundation and from the Peter J. Shields Endowment Fund for Dairy Food Science at the University of Cali- fornia Davis. V.P.D.-C. and B.L. received stipends from the Nestlé Nutrition Institute for participation in the 94th NNI workshop. D.A.M. has no financial relationship to disclose.

References

1 Bunker JJ, Erickson SA, Flynn TM, et al: Natural 13 Mathias A, Corthésy B: Recognition of gram-pos- polyreactive IgA antibodies coat the intestinal itive intestinal bacteria by hybridoma- and colos-

microbiota. Science 2017; 358:eaan6619. trum-derived secretory immunoglobulin A is

2 Mestecky J, Russell MW, Jackson S, Brown TA: mediated by carbohydrates. J Biol Chem 2011;

The human IgA system: a reassessment. Clin Im- 286: 17239–17247.

munol Immunopathol 1986; 40: 105–114. 14 Nakajima A, Vogelzang A, Maruya M, et al: IgA 3 Jarick KJ, Mokhtari Z, Scheller L, et al: Photocon- regulates the composition and metabolic function version of alloreactive T cells in murine Peyer’s of gut microbiota by promoting symbiosis be-

patches during acute graft-versus-host disease: tween bacteria. J Exp Med 2018; 215: 2019–2034. tracking the homing route of highly proliferative 15 Mathias A, Duc M, Favre L, et al: Potentiation of

cells in vivo. Front Immunol 2018; 9: 1468. polarized intestinal Caco-2 cell responsiveness to 4 Fagarasan S, Muramatsu M, Suzuki K, et al: Criti- probiotics complexed with secretory IgA. J Biol

cal roles of activation-induced cytidine deami- Chem 2010; 285: 33906-33913. nase in the homeostasis of gut flora. Science 16 Reikvam DH, Derrien M, Islam R, et al: Epitheli-

2002; 298: 1424–1427. al-microbial crosstalk in polymeric Ig receptor

5 Mazanec MB, Kaetzel CS, Lamm ME, et al: Intra- deficient mice. Eur J Immunol 2012; 42: 2959– cellular neutralization of virus by immunoglobu- 2970.

lin A antibodies. Proc Natl Acad Sci USA 1992; 89: 17 Rogier EW, Frantz AL, Bruno MEC, Kaetzel CS: 6901–6905. Secretory IgA is concentrated in the outer layer of 6 Moor K, Diard M, Sellin ME, et al: High-avidity colonic mucus along with gut bacteria. Pathogens

IgA protects the intestine by enchaining growing 2014; 3: 390–403.

bacteria. Nature 2017; 544: 498–502. 18 Donaldson GP, Ladinsky MS, Yu KB, et al: Gut 7 Forbes SJ, Martinelli D, Hsieh C, et al: Associa- microbiota utilize immunoglobulin A for muco-

tion of a protective monoclonal IgA with the O sal colonization. Science 2018; 360: 795–800. antigen of Salmonella enterica serovar 19 Corthesy B, Phalipon A: Molecular definition of Typhimurium impacts type 3 secretion and the role of secretory component in secretory IgA-

outer membrane integrity. Infect Immun 2012; mediated protection at mucosal surfaces. J Al-

80: 2454–2463. lergy Clin Immunol 2002; 109(1, suppl 1):S113. 8 Huang J, Guerrero A, Parker E, et al: Site-specific 20 Gibbins HL, Proctor GB, Yakubov GE, et al: SIgA glycosylation of secretory immunoglobulin A binding to mucosal surfaces is mediated by mu-

from human colostrum. J Proteome Res 2015; 14: cin-mucin interactions. PLoS One 2015; 1335–1349. 10:e0119677. 9 Perrier C, Sprenger N, Corthésy B: Glycans on 21 Torow N, Marsland BJ, Hornef MW, et al: Neona- secretory component participate in innate protec- tal mucosal immunology. Mucosal Immunol

tion against mucosal pathogens. J Biol Chem 2017; 10: 5–17.

2006; 281: 14280–14287. 22 Hanson LA: Breastfeeding provides passive and 10 Dallas SD, Rolfe RD: Binding of Clostridium dif- likely long-lasting active immunity. Ann Allergy

ficile toxin A to human milk secretory compo- Asthma Immunol 1998; 81: 523–533; quiz 533–

nent. J Med Microbiol 1998; 47: 879–888. 534, 537. 11 Rochereau N, Drocourt D, Perouzel E, et al: Dec- 23 Tuaillon E, Valea D, Becquart P, et al: Human tin-1 is essential for reverse transcytosis of glyco- milk-derived B cells: a highly activated switched sylated SIgA-antigen complexes by intestinal M memory cell population primed to secrete anti-

cells. PLoS Biol 2013; 11:e1001658. bodies. J Immunol 2009; 182: 7155–7162. 12 Baumann J, Park CG, Mantis NJ: Recognition of secretory IgA by DC-SIGN: implications for immune surveillance in the intestine. Immunol Lett

2010; 131: 59–66.

Milk SIgA and Commensals 167

Ogra PL, Walker WA, Lönnerdal B (eds): Milk, Mucosal Immunity and the Microbiome: Impact on the Neonate. Nestlé Nutr Inst Workshop Ser. Basel, Karger, 2020, vol 94, pp 158–168 (DOI:10.1159/000505335) 24 Seppo AE, Savilahti EM, Berin MC, et al: Breast 30 Vassilev TL, Veleva KV: Natural polyreactive IgA milk IgA to foods has different epitope specificity and IgM autoantibodies in human colostrum.

than serum IgA – evidence for entero-mammary Scand J Immunol 1996; 44: 535–539.

link for food-specific IgA? Clin Exp Allergy 2017; 31 Planer JD, Peng Y, Kau AL, et al: Development of

47: 1275–1284. the gut microbiota and mucosal IgA responses in

25 Roux ME, McWilliams M, Phillips-Quagliata JM, twins and gnotobiotic mice. Nature 2016; 534: et al: Origin of IgA-secreting plasma cells in the 263–266.

mammary gland. J Exp Med 1977; 146: 1311– 32 Kau AL, Planer JD, Liu J, et al: Functional charac- 1322. terization of IgA-targeted bacterial taxa from un- 26 Nathavitharana KA, Catty D, McNeish AS: IgA dernourished Malawian children that produce antibodies in human milk: epidemiological mark- diet-dependent enteropathy. Sci Transl Med

ers of previous infections? Arch Dis Child Fetal 2015; 7: 276ra24.

Neonatal Ed 1994; 71:F192–F197. 33 Bunker JJ, Bendelac A: IgA responses to micro-

27 Harris NL, Spoerri I, Schopfer JF, et al: Mecha- biota. Immunity 2018; 49: 211–224. nisms of neonatal mucosal antibody protection. J 34 Gopalakrishna KP, Macadangdang BR, Rogers

Immunol 2006; 177: 6256–6262. MB, et al: Maternal IgA protects against the de- 28 Torres O, Cruz JR: Protection against Campylo- velopment of necrotizing enterocolitis in preterm

bacter diarrhea: role of milk IgA antibodies infants. Nat Med 2019; 25: 1110–1115. against bacterial surface antigens. Acta Paediatr 35 Rogier EW, Frantz AL, Bruno MEC, et al: Secre-

1993; 82: 835–838. tory antibodies in breast milk promote long-term 29 Dolby JM, Newell DG: The protection of infant intestinal homeostasis by regulating the gut mi- mice from colonization with Campylobacter crobiota and host gene expression. Proc Natl

jejuni by vaccination of the dams. J Hyg (Lond) Acad Sci USA 2014; 111: 3074–3079.

1986; 96: 143–151. 36 Davidson LA, Lönnerdal B: Persistence of human milk proteins in the breast-fed infant. Acta

Paediatr Scand 1987; 76: 733–740.

168 Dunne-Castagna/Mills/Lönnerdal