If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The goal of this study was to review the role of human milk in shaping the infant intestinal microbiota and the potential of human milk bioactive molecules to reverse trends of increasing intestinal dysbiosis and dysbiosis-associated diseases.
Methods
This narrative review was based on recent and historic literature.
Findings
Human milk immunoglobulins, oligosaccharides, lactoferrin, lysozyme, milk fat globule membranes, and bile salt–stimulating lipase are complex multifunctional bioactive molecules that, among other important functions, shape the composition of the infant intestinal microbiota.
Implications
The co-evolution of human milk components and human milk–consuming commensal anaerobes many thousands of years ago resulted in a stable low-diversity infant microbiota. Over the past century, the introduction of antibiotics and modern hygiene practices plus changes in the care of newborns have led to significant alterations in the intestinal microbiota, with associated increases in risk of dysbiosis-associated disease. A better understanding of mechanisms by which human milk shapes the intestinal microbiota of the infant during a vulnerable period of development of the immune system is needed to alter the current trajectory and decrease intestinal dysbiosis and associated diseases.
Whether shaped by the tremendous pressures of evolution or an incredible feat of bioengineering (or both), human milk contains not only a near-perfect balance of macronutrients and micronutrients for term infants but also a dizzying array of bioactive molecules to protect the infant against infection, shape the intestinal microbiota, and tune the developing innate and adaptive immune systems of the infant. The abandonment of breastfeeding in favor of the convenience of formula feeding in the middle decades of the 20th century is one of several factors leading to an increase in intestinal dysbiosis (Table I) and its associated diseases (Table II). Attempts to improve infant formulas with the addition of components similar to those found in human milk have met with only limited success to date.
Table IFactors contributing to intestinal dysbiosis.
Table IIDysbiosis-associated diseases (all associations are between intestinal dysbiosis and the noted disease except early-onset sepsis and Group B streptococcal [GBS] sepsis in which the association is with maternal vaginal dysbiosis or GBS colonization of the genitourinary tract).
The purpose of the present review was to briefly summarize the evolution of human milk and the evidence that intestinal dysbiosis is part of the pathogenesis of several inflammatory diseases. We also describe how milk components have the potential to restore the infant microbiome, as well as the progress and future directions of bioengineering infant nutrition to decrease intestinal dysbiosis and dysbiosis-associated diseases. It is worth nothing that this is a narrative rather than a systematic review and does not touch on all bioactive aspects of human milk. The focus is on those milk molecules for which strong evidence suggests a role in shaping the microbiota; thus, human milk hormones, cytokines, growth factors, exosomes, and microRNAs are not discussed.
The History of Human Milk
Providing nutrition to the immature offspring is essential to the survival of a species. In the human, the placenta is the primary source of both nutrition and immunoglobulins for the fetus, complemented by swallowed amniotic fluid, which is rich in nutrients and bioactive molecules. Following birth, there is a paradigm shift in both nutrition and immune protection, with milk filling both roles for the first months of life (Figure 1). Depending on your viewpoint, human milk has either evolved over millions of years to maximize chances of survival of the infant at significant cost to the mother or was the most complex and complete nutritional creation of all time.
Figure 1Prenatal (A) and postnatal (B) transfer from mother to offspring of nutrients; immunologic, antimicrobial, and growth factors; and microbes. Panel A displays transplacental passage of nutrients and immunoglobulin G (IgG; blue) providing nutrition and immunity to the fetus. Swallowed amniotic fluid is rich in growth factors promoting fetal intestinal development and contains nutrients. During vaginal birth, the infant is exposed to maternal microbes. Panel B, Postnatally, breast milk provides secretory immunoglobulin A (SIgA), nutrients, microbes, and several immune, antimicrobial, and growth factors. Skin-to-skin care exposes the infant to maternal skin microbes. Copyright Satyan Lakshminrusimha.
Lactation and the consumption of milk are the defining characteristics of mammals; clues to its origin can be found by studying the evolution of mammary glands,
The predecessors to mammals likely appeared >300 million years ago, with mammals, hominids, and Homo sapiens appearing about 178 million, 4 million, and 150 thousand years ago, respectively. Comparisons between human milk and the milk of non-human primates exhibit dramatic differences in oligosaccharide composition
suggesting that changes in lactation were central to the emergence and survival of our species with its large brains and vulnerable offspring. Unfortunately, the fossil record provides no information regarding differences in milk composition between Homo sapiens and earlier hominids.
Lacto-engineering refers to tailoring milk to the needs of the infant. It is most commonly used to describe interventions to alter the composition of human milk such as feeding of hindmilk to increase fat intake or addition of fortifiers rich in macronutrients and micronutrients. In this paper, we broaden the term to include tailoring milk, either in the breast or after expression, to alter the intestinal microbiota of the infant.
Intestinal Dysbiosis
Bacteria were among the first living species on earth and have existed and evolved for several hundred millions to several billions of years. The primary drivers of bacterial evolution include the capacity to survive in a given location or niche; examples include defense mechanisms, virulence factors, and, perhaps most importantly, competition for nutrients. The emergence of a subspecies of Bifidobacterium with a unique set of genes encoding transport proteins and glycosidases necessary for the complete digestion of the dozens of unique human milk oligosaccharides at roughly the same time as the appearance of Homo sapiens is a compelling example of co-evolution of 2 interdependent species.
For most of human history, mothers have provided an array of human milk oligosaccharides and other bioactive molecules to shape the intestinal microbiota of their infants. The few bacteria capable of consuming human milk oligosaccharides (select Bifidobacterium and Bacteroides species) thus have had a distinct advantage in colonizing the neonatal intestinal tract. This pattern of a stable infant community with low diversity and anti-inflammatory properties differs dramatically from the intestinal microbiota of the adult and plays an important role in instructing and shaping both the innate and the adaptive immune systems.
Layered immunity' and the 'neonatal window of opportunity'—timed succession of non-redundant phases to establish mucosal host-microbial homeostasis after birth.
Dysbiosis refers to alterations in the microbiota of a given niche associated with disease. In humans, the most common examples include Clostridium difficile colitis and antibiotic-associated diarrhea. In both cases, exposure to antibiotics alters the intestinal microbiota, creating an opportunity for pathobionts to flourish, which leads to inflammation that may be acute or chronic. The recent descriptions of associations between alterations in the intestinal microbiota and a wide variety of autoimmune and acute and chronic inflammatory diseases (Table II) have challenged assumptions about the origins and pathogenesis of many diseases that have been increasing over the past several decades.
It seems likely that significant changes in hygiene, diet, exposure to antibiotics (both prescribed and in foods), and many life-saving medical advances (eg, cesarean delivery) have had the unexpected consequence of fundamentally altering the intestinal microbiota and that these changes now span several generations such that mothers delivering infants today are not colonized with and therefore unable to transfer many bacterial species that colonized our ancestors (Figure 2).
Mechanisms by which gut microbes influence the developing immune system include direct impacts on T helper type 1/T helper type 2/regulatory T cell (Treg) balance,
Indole-3-lactic acid, a metabolite of tryptophan, secreted by Bifidobacterium longum subspecies infantis is anti-inflammatory in the immature intestine.
Figure 2Role of breastfeeding in establishing an intestinal microbiota rich in Bifidobacterium and Bacteroides species. Factors that can disrupt this process such as delivery by cesarean section, antibiotic therapy, and formula feeding are shown. Secretory immunoglobulin A (SIgA) in breast milk provides local, mucosal immunity in the intestine. Immunoglobulin G (IgG) antibodies in the breast milk may be absorbed to provide systemic immunity. Breast milk can also enhance memory lymphocytes to reside in the Peyer's patches, leading to long-standing immunity against enteral pathogens. FcRn = neonatal Fc receptor; hMO = human milk oligosaccharide. Copyright Satyan Lakshminrusimha.
Human Milk Components That Shape the Infant Intestinal Microbiota
Immunoglobulins
Neonates are born immunologically immature and susceptible into an environment abundant in pathogens. This susceptibility is evident in the high rates of infant mortality from diarrheal and respiratory infections and has been attributed to immature innate and adaptive immunity, as well as a deficiency in diversity of the commensal microbiota.
Transfer of maternal antibodies by milk provides protection against pathogens until immune system maturation, particularly against organisms that gain access via mucosal barriers (Figure 3).
Figure 3Components of breast milk with immunomodulatory function that are harmed by pasteurization (blue boxes) or unharmed by pasteurization (red boxes). SIgA = secretory immunoglobulin A. Copyright Satyan Lakshminrusimha.
Breast milk antibodies are predominantly made up of secretory immunoglobulin A (SIgA) (>90%) but also consist of immunoglobulin M and immunoglobulin G (IgG). SIgA (and immunoglobulin M, to a lesser extent) work locally by protecting the neonatal mucosal surfaces through mechanisms such as toxin or adhesin neutralization, bacterial agglutination, and enchained growth,
ultimately physically impeding bacterial interactions with the gut tissues (“immune exclusion”) and thus preventing inflammation (Figure 2). This ability for exclusion of luminal bacteria is not critically dependent on antibody specificity.
Interactions between milk SIgA and a subclass of Tregs may be of particular importance. Tregs limit inflammation but delay pathogen clearance. Although they are important during initial colonization of the neonate, mechanisms regarding regulation of Tregs are unclear. In mice, Tregs are induced by the intestinal microbiota around day 14 of life (early weaning), but maternal milk SIgA influenced Treg numbers much earlier (the first 3 days of life), and this impact persisted over the lifetime of the mouse and even into subsequent generations. Milk SIgA seems to create a set-point for Tregs that is inherited from mother to pup in a nongenetic, nonepigenetic, nonmicrobe mechanism with long-term multigenerational consequences.
The mother's intestinal microbiota influences production of plasma cells in the Peyer's patches of the maternal small intestine; these plasma cells, programmed by maternal gut microbes, secrete SIgA into the milk.
in whom IgA absorption is believed to be secondary to greater passive intestinal permeability. In addition, neonatal IgA production is enhanced from maternally derived microbiota and can be transported to mucosal secretions, although evidence to support a protective role of this particular IgA against common enteric pathogens compared with SIgA found in breast milk is lacking.
The impact of milk SIgA on the infant intestinal microbiota is supported by the following observations: individuals and animals deficient in IgA have an increased relative abundance of Enterobacteriaceae in their feces, and IgA appears to support the growth of obligate anaerobes such as Bacteroides in the gut.
Provision of maternal SIgA in breast milk not only altered the intestinal microbiota of the pups but resulted in alterations of intestinal epithelial cell gene expression that persisted into adulthood.
(Figure 2). The neonatal Fc receptor expressed on intestinal epithelial cells mediates the neonatal uptake of maternal-derived IgG from ingested milk into the small intestine, which is the same receptor that aids transport of maternal IgG across the placenta. In the mouse, maternal pathogen-specific IgG in the neonatal gut lumen, but not IgA or immunoglobulin M, plays a critical role in neonatal protection.
In this model, protective immunity against enteric infection was induced after oral infection but not parenteral immunization, correlating with the presence of pathogen-specific IgA and IgG in the intestinal lumen. Both oral and parenteral immunization of dams protected offspring against oral pathogen challenge, correlating with lower pathogen loads in feces, liver, and spleen of the pups as well as reduced inflammation and mortality. Maternal IgG opsonizes pathogens, increases phagocytosis by neutrophils, and reduces pathogen attachment to the intestinal epithelium during the neonatal period. Long-lasting immunologic memory relies on the generation of memory B cells in the germinal centers of intestinal Peyer's patches and lymphoid follicles, which differentiate into plasma cells that produce mucosal antibodies in the gut.
Maternal breast milk–derived IgG also plays an important role in establishing intestinal immune homeostasis after microbial colonization and exposure to commensal antigens. Maternal antibody isotypes IgG2b and IgG3 in breast milk are important in promoting functional mutualism with the developing neonatal intestinal microbiota by limiting mucosal adaptive immune responses through dampening the development of T follicular helper cells and subsequent germinal center responses against intestinal microorganisms.
These antibody responses are T-cell independent and largely Toll-like receptor (TLR) dependent. This modulation of the interaction between the microbiota and intestinal mucosa appears to play an important role in the prevention of necrotizing enterocolitis (NEC).
Maternal natural antibodies (mNabs) are antibodies acquired without known exposure to the pathogen or through immunization. Using a murine model, Zheng et al
showed how a single commensal species can induce cross-reactive mNabs that protect against pathogens. These IgG mNabs can be assimilated directly from ingested milk into serum by a neonatal Fc receptor–dependent process. IgG binding to the microbial surface and entering the bloodstream can additionally drive immune-effector functions such as complement-dependent bacteriolysis and opsonization, protecting against invasive pathogenic infections.
Future Directions
Further research is needed to understand the mechanisms underlying milk immunoglobulin production (eg, how do alterations in the maternal gut microbiota, maternal disease processes and maternal genetics/diet/medications influence milk SIgA levels?), the impact of milk SIgA on the milk microbiota, mechanisms by which milk SIgA shape the infant gut microbiota, and the importance of antibodies found in relatively low abundance (eg, specific high-affinity IgG antibodies, mNabs). Standard pasteurization denatures milk immunoglobulins; development of novel methods to ensure safety of human milk without damaging antibodies would improve the protective effects of donor milk. Immunization during pregnancy is currently limited to influenza, pertussis, and coronavirus disease 2019, with strong evidence that transfer of antibodies through milk is effective at preventing disease in the infant for influenza and pertussis. Novel vaccines against common causes of diarrhea and respiratory infections that could be safely given during pregnancy and lactation may be beneficial in decreasing morbidity and mortality in developing countries. Clinical trials of these vaccines should include analysis of milk immunoglobulins and the infant gut microbiota.
Human Milk Oligosaccharides
All mammals produce oligosaccharides of varying structure, diversity, and size, with humans exhibiting the highest degree of complexity.
commonly divided into categories (fucosylated, sialylated, or nondecorated [meaning no fucose or sialic acid]). hMOs are diverse, as well as highly abundant, representing a greater volume in human milk than protein. Although there is variation between women and across geographic regions, hMO abundance is 9 to 22 g/L in colostrum, 8 to 19 g/L in transitional milk, 6 to 15 g/L at 1 month of lactation, and 4 to 6 g/L after 6 months
; however, the human infant is unable to digest these oligosaccharides to utilize them as an energy source. The compelling question has been, why does a mother expend tremendous energy to produce a wide variety and large volume of hMOs at great cost to herself when these relatively small sugar molecules are not digestible by her infant? Analyses involving culture media in which hMOs are the only carbon source have shown that very few species of bacteria that inhabit the human gut are able to transport or digest hMOs (mostly limited to just a few species and subspecies of Bifidobacterium and Bacteroides).
Selective utilization of the human milk oligosaccharides 2′-fucosyllactose, 3-fucosyllactose, and difucosyllactose by various probiotic and pathogenic bacteria.
Thus, the primary purpose of hMOs in mother's milk appears to be to provide a selective advantage to a few commensal bacteria to shape the intestinal microbiota of the infant.
This hypothesis is supported by the following direct observations: (1) increased total hMOs in milk are associated with increased abundance of Bifidobacterium species in the milk, with specific categories of hMOs influencing abundance of individual species
Human milk glycomics and gut microbial genomics in infant feces show a correlation between human milk oligosaccharides and gut microbiota: a proof-of-concept study.
; and (4) hMO composition is a risk factor for stunting and malnutrition in infants, with causality shown in animal models between specific hMO structures, the intestinal microbiota, and growth.
In addition to shaping the infant microbiota with resulting effects on the developing immune system, hMOs appear to have direct antimicrobial effects and to influence gut maturation and barrier function.
A recent review summarized several pathways underlying the immunomodulatory effects of hMOs and their potential role in prevention of allergic diseases early in life.
Three strategies for imitating the prebiotic effects of hMOs have been explored. Synthetic prebiotic glycans such as inulin, galacto-oligosaccharides, and fructo-oligosaccharides have been added to infant formula or administered directly to infants in hopes of altering the intestinal microbiota, improving feeding tolerance, and decreasing the risk of sepsis and NEC. A recent meta-analysis of 18 randomized controlled trials (1322 preterm infants) found a decrease in sepsis, mortality, and feeding intolerance (but not NEC) with prebiotic administration.
Effects of prebiotics on sepsis, necrotizing enterocolitis, mortality, feeding intolerance, time to full enteral feeding, length of hospital stay, and stool frequency in preterm infants: a meta-analysis.
Second, a limited number of hMO structures (eg, 2′-fucosyllactose, lacto-N-neotetraose) have been added to term infant formula, with small studies showing lower levels of cytokines and fewer episodes of respiratory infection and alteration in the fecal microbiota,
Similar to those who are breastfed, infants fed a formula containing 2′-fucosyllactose have lower inflammatory cytokines in a randomized controlled trial.
Finally, a mixture of bovine milk oligosaccharides added to infant formula in a randomized trial exhibited increased relative abundance of fecal Bifidobacterium, decreased relative abundance of fecal Clostridium, and increased fecal IgA compared with a control formula without added oligosaccharides.
Term infant formula supplemented with milk-derived oligosaccharides shifts the gut microbiota closer to that of human milk-fed infants and improves intestinal immune defense: a randomized controlled trial.
Further research is needed to understand differences in the impact of specific hMO structures (eg, are some hMOs more important in shaping the microbiota or influencing gut maturation or inflammation than others?), the impact of hMOs that are absorbed from the gut (eg, do hMOs in the bloodstream and urinary tract play an immunologic or antibacterial role?), and the importance of genetic differences in hMO production in preventing intestinal dysbiosis (eg, how do common mutations in fucosyltransferases affect the milk microbiota, the intestinal microbiota, and infant susceptibility to disease?; is supplementation of specific hMOs to infants of mothers who are unable to secrete those hMOs protective?). As a wider array of milk oligosaccharides (either synthetic or from animal sources) becomes available, studies of combinations of these glycans for the prevention of dysbiosis-associated diseases have high potential.
Lactoferrin
Lactoferrin (LF), the second most abundant protein in breast milk,
is a multifunctional bioactive glycoprotein. It is known to have antimicrobial, anti-inflammatory, immunomodulatory, and growth-promoting properties that may prevent bacterial translocation in the intestines of premature neonates.
LF exerts its antimicrobial and antifungal effects through a variety of mechanisms. It is bacteriostatic by binding to iron and making it inaccessible to pathogens in the intestinal lumen that require iron to proliferate. It also binds to lipopolysaccharide (LPS) and lipoteichoic acid on the cell surface, disrupting the bacterial cell membrane and decreasing biofilm formation.
Several studies have shown potent bactericidal activity against pathogens as a result of formation of lactoferricin, a peptide released during digestion of LF.
Several translational investigators of LF prophylaxis have reported improved bacterial clearance and survival in mice and piglets receiving LF prophylaxis.
binding to LPS (resulting in reduction in upregulation of inflammatory cytokines), and decreasing the production of tumor necrosis factor and other pro-inflammatory cytokines.
LF promotes growth of beneficial bacteria, specifically Lactobacillus and Bifidobacterium, which, in turn, can limit the growth of pathogens by decreasing intestinal pH.
found a significant reduction of late-onset sepsis, NEC, and death in preterm infants receiving prophylactic bovine LF compared with placebo, with the most pronounced impact seen in preterm infants with birth weight <1000 g. In a secondary analysis, the group found a significant decrease in invasive fungal infections.
The effect of lactoferrin supplementation on death or major morbidity in very low birthweight infants (LIFT): a multicentre, double-blind, randomised controlled trial.
in Australia and New Zealand also showed no difference in mortality or major morbidity (including NEC); however, the group performed a Cochrane meta-analysis of >5000 infants, which found that although there was no improvement in death or major morbidity, LF did reduce late-onset sepsis.
A recent nested study of a subset of the infants in the ELFIN trial showed that LF altered the fecal microbiota compared with matched control infants (with decreases in fecal Staphylococcus and other potential pathogens); however, the changes were not as large as those associated with other clinical variables. In addition, LF administration did not appear to alter the infant metabolome.
Further research is needed to understand the impact of digestion (partial digestion releases the highly active peptide lactoferricin, and yet further digestion reduces bioactivity; is the limited capacity for proteolysis in the preterm infant advantageous in this regard?) and absorption of LF (does absorption of intact LF or lactoferricin result in significant tissue concentration in infection? is a leakier intestinal mucosa, as seen in the preterm infant, advantageous?), the antiviral effects of LF, the impact of LF on the milk microbiota, and the synergistic effects of LF with other bioactive milk molecules. More detailed study of the impact of glycosylation on the antimicrobial and immunostimulatory effects of LF is needed. Further clinical trials of LF in preterm infants (beyond those in progress) may be difficult to justify; however, it is possible that LF in combination with other human milk bioactive molecules could alter the intestinal microbiota and help decrease the risk of NEC and late-onset sepsis in preterm infants and dysbiosis-associated disease in term infants.
Lysozyme
Lysozyme is a ubiquitous antibacterial enzyme abundant in the granules of macrophages, Paneth cells, and neutrophils and in many secretions, including mucus, amniotic fluid, and breast milk.
Lysozymes enact their antimicrobial properties through the degradation of the outer wall of gram-positive bacteria, hydrolyzing β-1,4 bonds between N-acetylmuramic acid and N-acetyl-d-glucosamine in the peptidoglycan layer.
They also have bactericidal activity against gram-negative bacteria in vitro, acting synergistically with LF, which binds to the LPS in the outer bacterial membrane, removing it and allowing lysozyme to access and degrade the internal proteoglycan matrix of the membrane, thereby killing the bacteria.
These mechanisms are also important in preventing bacterial binding to intestinal epithelial cell receptors and therefore bacterial translocation.
Most strains of infant-type human-residential bifidobacteria (HRB) can grow and survive in human milk, although the capacity to do so appears to be strain dependent and breast milk dependent.
Infant-type HRB are tolerant of high lysozyme concentrations, correlating well with growth in breast milk; however, adult-type HRB and non-HRB were generally susceptible to lysozymes and therefore unable to grow in breast milk, suggesting that lysozyme in human milk could selectively exclude colonization of non-HRB in the intestines of breastfed infants. The mechanism behind this differential tolerance is still unknown.
Pigs consuming transgenic goat milk containing human lysozyme (HLZ) had reduced clinical severity and improved recovery from enterotoxigenic Escherichia coli infection,
Human lysozyme expressed in the mammary gland of transgenic dairy goats can inhibit the growth of bacteria that cause mastitis and the cold-spoilage of milk.
; TLR4 binds LPS from gram-negative bacteria and stimulates the innate immune system, whereas interleukin-10 is an anti-inflammatory cytokine produced by monocytes and regulatory T cells that inhibits LPS-induced stimulation of inflammatory cytokines. Oral administration of hen egg white lysozyme in a porcine model of chemically induced colitis led to decreased expression of the pro-inflammatory cytokines tumor necrosis factor-α and interleukin-8, as well as increased expression of anti-inflammatory transforming growth factor-β1.
observed similar findings on transforming growth factor-β1 expression in the ileum of pigs fed goat's milk containing HLZ. Interestingly, transgenic HLZ goat milk also has reduced spoilage compared with milk from nontransgenic control goats, which could be of benefit to regions with poor access to refrigeration.
Human lysozyme expressed in the mammary gland of transgenic dairy goats can inhibit the growth of bacteria that cause mastitis and the cold-spoilage of milk.
Both HLZ and human LF have been expressed in rice, a common food for infants. A randomized trial of oral rehydration solution containing both human rice proteins administered to infants with acute diarrhea and dehydration showed shorter duration of diarrhea compared with standard oral rehydration solutions.
As noted earlier, lysozymes are also produced by macrophages, neutrophils, and Paneth cells (secretory cells at the base of the small intestinal crypts that protect stem cells and shape the intestinal microbiota). Paneth cell immaturity in the premature infant may play an important role in the pathogenesis of NEC. Infants with NEC have fewer Paneth cells than matched controls, and a Paneth cell ablation mouse model generates a NEC-like phenotype.
It remains uncertain which of the many antimicrobial proteins and peptides secreted by Paneth cells are most important in shaping the microbiota of the small intestine; however, given that Paneth cells are not fully functional until at or following term gestation, it is likely that this important function (influencing the composition and function of the intestinal microbiota) is performed sequentially by swallowed amniotic fluid, human milk, and then Paneth cells, all containing lysozyme. No clinical trials, however, have yet to assess the benefits of lysozyme supplementation in the neonatal population.
Future Directions
Further research to understand the selective effects of lysozyme on bifidobacteria and other commensal gut microbes is needed (how do select commensal bacteria resist killing by lysozyme, and why has this apparent tolerance factor not been acquired by pathogens?). Further study of the impact of combined milk bioactives (eg, lysozyme plus LF) on the microbiota and intestinal inflammation and permeability would be of value in preparation for clinical trials. The milk of transgenic livestock delivering HLZ has the potential to decrease dysbiosis-associated diseases, including diarrheal diseases in developing countries; given the substantial mortality of diarrhea in infants, clinical trials of this approach are certainly justified.
Milk Fat Globule Membrane
Milk is essentially an emulsification of milk fat globules (MFGs) in an aqueous solution. MFGs are bags of fat of varying sizes (range of 0.2 to 15 μm in human milk). They consist of a membrane surrounding a core of triacylglycerols that provide about one half of human milk's caloric content.
The membrane is a tri-layer with an inner polar lipid monolayer derived from the endoplasmic reticulum, a central protein-rich layer, and an outer polar lipid bilayer derived from the apical membrane of the mammary epithelial cell. The MFG membrane maintains the stability of the MFG in the lipid-in-water emulsion and protects the triacylglycerols from enzymatic digestion and coalescence. This MFG membrane is biologically active with a variety of phospholipids, sphingolipids, glycolipids, proteins, glycoproteins, and cholesterols. In addition to nutritional, metabolic, and synthetic properties (eg, the MFG membrane contains the majority of choline-containing phospholipids in milk), the MFG membrane also has antimicrobial, anti-inflammatory, and anticancer properties.
MFG membranes have predominantly been studied in bovine milk; however, a recent summary of metabolomic studies showed both similarities and differences in MFG membrane proteins and glycoproteins across stages of lactation and among 5 species of mammals (bovine, goat, yak, buffalo, and human).
Comparative proteomics of milk fat globule membrane (MFGM) proteome across species and lactation stages and the potentials of MFGM fractions in infant formula preparation.
Comparative proteomics of milk fat globule membrane (MFGM) proteome across species and lactation stages and the potentials of MFGM fractions in infant formula preparation.
Pasteurization does not appear to alter the size distribution of MFGs; however, many of the MFG proteins are altered with significant differences between goat, bovine, and human samples.
A recent review of 6 double-blind, randomized controlled trials of infant formula with added MFG membrane found a positive effect on cognitive development in 2 of the studies and a protective effect against infection in 3 of the studies.
Fecal microbiome and metabolome of infants fed bovine MFGM supplemented formula or standard formula with breast-fed infants as reference: a randomized controlled trial.
Table IIIMilk fat globule membrane components. Listed are some of the most abundant components; those in bold are either antimicrobial or have been shown to influence the infant intestinal microbiota.
Butyrophilin
Glycerophospholipids (phosphatidylethanolamine, phosphatidylcholine, phosphatidylinositol, and phosphatidylserine)
Xanthine dehydrogenase/oxidoreductase
Adipophilin
Sphingolipids (sphingomyelin)
Glycosphingolipids (gangliosides, cerebrosides)
Lactadherin (PAS VI/VII, MFG-EGF factor 8)
Mucins (MUC1, MUC4, MUC15)
Cholesterol
Fatty acid synthase
Ras-related protein Rab-18
Glycomacropeptide (a peptide from digestion of kappa-casein)
Lactoferrin/lactotransferrin
Isracidin (a peptide from digestion of alpha-S1-casein)
Because there is only one study of the impact of the MFG membrane on the infant fecal microbiota, further research is needed on this important question, underscoring the value of including fecal microbiota analysis in clinical trials. The hypothesis that the MFG membrane could be useful as a “protecting encapsulating matrix for lactic acid bacteria” has been proposed and is deserving of investigation.
The ready availability of commercial bovine and caprine MFG membrane and preliminary data from clinical trials of infant formulas with added MFG membrane suggest that further clinical trials in term and preterm infants are warranted.
Bile Salt–Stimulated Lipase
At birth, the enzymatic capacity of the gut of the term infant is immature. Production of gastric pepsin, pancreatic triglyceride lipase, bile salts, and phospholipase A2 are all low in the first weeks and months of life, a time of rapid brain growth and high metabolic demand. Digestion of human MFGs begins in the stomach with gastric lipase, which is similar in abundance and activity to that of adults, and continues in the proximal small intestine through the actions of pancreatic lipase-related protein 2 and bile salt–stimulated lipase (BSSL). The latter is abundant in human milk and becomes functional in the small intestine after the partial digestion of MFGs in the stomach.
; second, BSSL is denatured by standard pasteurization techniques; and third, post-pyloric feeding limits the exposure of MFGs to gastric lipase and therefore limits the functional activity of BSSL. As a result, both of these common interventions (post-pyloric feeding and pasteurization of human milk) may lead to limited lipolysis and fat malabsorption.
BSSL is highly glycosylated with differences in glycosylation over time of lactation,
Recombinant human-milk bile-salt-stimulated lipase. Functional properties are retained in the absence of glycosylation and the unique proline-rich repeats.
A pilot study of recombinant human BSSL in preterm infants showed improved weight gain and increased absorption of docosahexaenoic acid and arachidonic acid but no effect on total fat absorption.
A Phase III randomized trial of the same product in 415 preterm infants found no improvement in growth velocity, except in a predefined subgroup of small-for-gestational-age infants. This study reported increased adverse events in the BSSL group.
The study of the intestinal virome and its impact on the microbiota, immune development, and risk of dysbiosis-associated diseases is in its infancy. Evidence to date suggests that viral colonization of the infant gut follows bacterial colonization and is influenced by human milk.
Further research into the impact of various forms of BSSL on the intestinal microbiota and virome is needed. The description of expression of recombinant human BSSL in transgenic cows raises the possibility of large-scale production
Unpasteurized human milk contains both microbes from the mother's skin (predominantly Staphylococcus, Streptococcus, Propionibacterium, and Corynebacterium) and the infant's mouth (Streptococcus and Rothia). In addition, limited studies suggest the possibility of translocation of intestinal bacteria from the maternal gut into the milk via the lymphatics.
Factors influencing the milk microbiota include mode of breastfeeding (pumped vs direct), antibiotic exposure, probiotic ingestion, delivery mode, parity, maternal body mass index, mastitis, and preterm delivery. Several studies have shown strong community-level associations between the milk microbiota and the infant fecal microbiota.
The earliest studies of the fecal microbiota of the breastfed infant were performed before the antibiotic era and found a predominance of bifidobacteria (the most effective consumers of hMOs).
Studies have shown that a single subspecies, Bifidobacterium longum subspecies infantis, encodes in its genome all of the transferases and glycosidases necessary to transport all HMO structures into its cytoplasm and completely digest these hMOs into monosaccharides.
Provision of a probiotic B infantis strain to breastfed infants for just 3 weeks beginning at day of life 7 led to prolonged colonization with this strain
showing the synergy between hMOs and this bacterial subspecies; we can find no other examples of sustained alteration of the microbiota by a brief period of probiotic administration. In the pre-antibiotic era, how B infantis arrived in the infant gut is uncertain, but it seems likely that the origin was the mother's intestinal microbiota, where this microbe would have been a minor component, with passage to the infant either through the fecal–oral route or the milk microbiota, or both. As noted earlier, several components of human milk shape the milk microbiota, including hMOs and lysozyme.
The human milk virome has recently been characterized. The major component consists of bacteriophages with differences in composition influenced by preterm birth, delivery mode, birth weight, and stage of lactation.
The impact of the milk virome on the infant fecal virome, metabolome, and bacterial composition remains unknown.
Future Directions
One of the challenges of determining the functional impact of the milk microbiota on the infant intestinal microbiota is that gastric acid and bile acids inhibit growth of many bacteria, limiting passage of these microbes beyond the stomach and proximal small bowel. The impact of these microbes on the developing immune system is likely limited as they tend not to reach the distal small bowel with its rich collection of immune cells. Research emphasis on milk microbes that are resistant to gastric acid and bile acids (eg, Bifidobacterium and Lactobacillus species) is needed, as is detailed exploration of the impact of the milk virome on the developing microbiota and immune system. Further research into the impact of the maternal exposome on the relevant components of the milk microbiota is needed. Although improved methods of pasteurization may help limit damage to milk proteins, peptides, and glycoproteins, these are not likely to spare the milk microbiota. Consequently, research into the impact of adding probiotic microbes to pasteurized donor milk is needed.
In addition to the human milk components discussed here, limited evidence for other components that influence the infant gut microbiota is summarized in Table IV.
Table IVAdditional human milk components that may alter the intestinal microbiota.
Use of batch culture and a two-stage continuous culture system to study the effect of supplemental alpha-lactalbumin and glycomacropeptide on mixed populations of human gut bacteria.
Lacto-Engineering Within the Mammary Gland (in mamma) to Alter the Infant Microbiota
Many women struggle to produce a sufficient quantity of milk. In addition, the content of macronutrients and bioactive molecules in the milk varies from woman to woman, from foremilk to hindmilk, and over the time of lactation. Increasing the quantity and quality of human milk production has great potential to improve not just infant growth and nutrition but also the intestinal microbiota and consequently the risk of dysbiosis-associated diseases. The term exposome refers to all environmental exposures that influence health. Maternal diseases (eg, obesity, diabetes), diet, phytochemicals, and galactogogues all influence milk production, growth and overgrowth of the infant, and the infant microbiota.
Human milk microbial community structure is relatively stable and related to variations in macronutrient and micronutrient intakes in healthy lactating women.
The lack of research and understanding of optimization of the nutritional status of the mother–infant dyad have recently been reviewed and represent a tremendous opportunity to improve long-term health of both mother and infant.
WHO. Healthy Eating during Pregnancy and Breastfeeding: Booklet for mothers. In: Regional Office for Europe: Nutrition and Food Security. (2001). p. 26.
but provides only limited guidance for mothers and infants with unique health needs or genetics. None of these guidelines include analysis of milk quantity, macronutrients, bioactive components, or the composition of the infant microbiome or metabolome. Although personalized medicine is making important in-roads in many aspects of medicine, the opportunities for progress in care of the mother–infant dyad are significant but largely ignored by industry and funding agencies. For instance, the current understanding of the safety and efficacy of galactogogues is limited; pharmaceuticals (eg, domperidone, metoclopramide) have limited efficacy and significant potential side effects, and botanicals (eg, fenugreek) have only minimal evidence regarding mechanisms of action, dosing, safety, and efficacy.
Oxytocin receptor agonists and nicotinamide derivatives offer promising alternatives for increasing milk production and quality, with early clinical trials in progress.
Maternal consumption of probiotics altered the infant fecal microbiota and the milk microbiota in several studies, and maternal consumption of lipid-based nutritional supplements increased bacterial diversity in the infant feces; maternal consumption of vitamin D or prebiotic supplements did not significantly alter the infant microbiota. A study of maternal fish oil consumption, not included in the previously noted meta-analysis, showed increased Bifidobacterium and Lactobacillus species in the infant feces.
Attempts to decrease risk of preterm delivery or increase infant birth weight by altering the maternal vaginal or intestinal microbiota with probiotics have been disappointing to date.
In a single small study, maternal consumption of probiotic dietary supplements after birth did not decrease the risk of NEC or mortality in preterm infants
; however, a meta-analysis of maternal consumption of probiotics during pregnancy and lactation exhibited a significant reduction in atopic dermatitis but not other allergic diseases,
A full review of traditional botanicals and phytochemicals that influence lactation and the infant gut microbiota is beyond the scope of this article and would be a valuable addition to the literature.
Lacto-Engineering Expressed Human Milk to Alter the Infant Microbiota
Currently, approaches to fortify expressed human milk (either from the infant's own mother or from a donor) are limited to the addition of fortifiers derived from bovine or donor human milk. To date, attempts to produce components of human milk on a large scale have focused on extracting similar or equivalent components from bovine milk,
These approaches have fostered the fortification of infant formula with milk oligosaccharides and MFG membrane for term infants, but the addition of milk bioactives to expressed human milk currently remains in the realm of research. This is due in part to the lack of capacity to measure the content of promising milk bioactive molecules in expressed milk or any functional aspects of the infant microbiota in the clinical setting. A milk analyzer that could measure hMOs, immunoglobulins, and LF and a bedside test to detect fecal dysbiosis would be major steps forward in the early identification of mother–infant dyads at particularly high risk for dysbiosis-related disease.
Two additional challenges to altering human milk (or formula) to improve the infant microbiota are limited capacity to synthesize more complex structures (eg, SIgA and larger hMOs) and limited posttranslational modification (eg, glycosylation of bioactive proteins and peptides). In January 2020, 108Labs produced the first cell-cultured human milk using primary human mammary cells in three-dimensional hollow fiber bioreactors,
since announcing their own cell-cultured milks. Cell-cultured human milk bioreactors produce a comprehensive range of human milk proteins, lipids, and hMOs simultaneously. Biosynthesis of human proteins in lactating primary human mammary cells may yield therapeutic proteins with more normal posttranscriptional modifications and higher quality glycans, with a higher possibility of N-linked, fucosylated, and large O-linked oligosaccharides shown to positively affect in vivo bioactivity and stability.
The failure of recombinant milk proteins in previous human trials may be partially attributable to abnormal posttranscriptional modifications and lesser quality glycan patterns typical of recombinant proteins compared with native proteins.
(Figure 4). Cell-cultured human SIgA administered with or without other therapeutic milk proteins and hMO derived from ex vivo lactating human mammary cells could be added to expressed mother's milk, pasteurized donor human milk, or infant formula to treat or prevent intestinal dysbiosis in preterm infants and/or term infants at high risk for dysbiosis-associated disease.
Figure 4Electron microscopy of 108Labs' first batch of native human secretory immunoglobulin A dimers biosynthesized on January 20, 2020, in 108Labs' secretory immunoglobulin A bioreactor, isolated from supernatant by affinity resin and diluted to 0.005 μg/mL for single protein imaging.
Although the American Academy of Pediatrics recommends pasteurized donor human milk for preterm infants whose mothers are unable to produce a sufficient supply of their own milk, and the US Food and Drug Administration recommends against obtaining donor milk from sources that do not screen donors or ensure the safety of donor milk,
It is likely that cell-cultured human milk–derived therapeutic proteins will be regulated and marketed exclusively as biologic drugs. Oversight of this process through the Investigational New Drug process by the US Food and Drug Administration requires a significant investment of time and resources but is essential to ensuring safety and efficacy.
Conclusions
Human milk is a complex tissue containing macronutrients and micronutrients, bioactive cells and molecules, and potentially beneficial microbes that nourish and protect the fragile infant during a period of rapid growth and high vulnerability to infectious diseases. Immunoglobulins, hMOs, LF, lysozyme, and human MFG membranes not only protect the infant from infections but help shape the infant intestinal microbiota, potentially protecting the infant against a long list of intestinal dysbiosis-associated diseases. Novel methods of pasteurization that destroy pathogens but leave bioactive molecules intact would improve the protective benefits of donor milk. Methods of boosting specific bioactive molecules in milk (eg, maternal vaccination during pregnancy or immediately after preterm birth, dietary supplements, and novel galactogogues) are promising. Bioengineering some or several of the bioactive molecules described herein as additives to mother's milk (eg, α1,2 fucosylated hMOs for mothers with a common mutation that precludes production of these hMOs) or to donor milk or infant formulas may be helpful in bridging the gaps between human milk–fed and formula-fed infants in partially restoring our ancestral infant microbiota and in reducing risk for dysbiosis-associated disease.
Acknowledgments
The National Institutes of Health have provided grant support to Dr. Lakshminrusimha (R01 HD072929) and Dr. Underwood (R21 HD096247).
Dr. Zeinali wrote the initial draft of several sections and edited the entire manuscript. Dr. Giuliano wrote the sections on three-dimensional mammary cell culture, biosynthesis of immunoglobulins, and other bioactive human milk molecules. Dr. Lakshminrusimha created Figures 1 to 3 and edited the manuscript. Dr. Underwood wrote the initial draft of several sections and edited the entire manuscript. All authors approved the final version as submitted.
Disclosures
Dr. Giuliano is a co-founder and the CEO of 108Labs. Dr. Underwood has received an honorarium from Prolacta for speaking at conferences on probiotics. Drs Zeinali and Lakshminrusimha have indicated that they have no conflicts of interest regarding the content of this article.
Supplementary materials
A co-culture of human primary mammary epithelial and B cells were grown and serial passaged for 30 days using mammary growth medium. The cells were then seeded into a hollow fiber bioreactor and cultured with growth medium for 10 more days until glucose consumption stabilization was observed. Lactation medium was then used to stimulate milk production and LLPC niche function, corresponding with an observed increase in the production of secretory IgA as measured by SIgA ELISA. Supernatant was harvested from the bioreactor 25 days after seeding. Secretory antibodies were isolated from supernatant using an affinity resin, and directly imaged with electron microscopy at 0.005 ug/mL.
References
Schep R.
et al.
Control of Hoxd gene transcription in the mammary bud by hijacking a preexisting regulatory landscape.
Layered immunity' and the 'neonatal window of opportunity'—timed succession of non-redundant phases to establish mucosal host-microbial homeostasis after birth.
Indole-3-lactic acid, a metabolite of tryptophan, secreted by Bifidobacterium longum subspecies infantis is anti-inflammatory in the immature intestine.
Selective utilization of the human milk oligosaccharides 2′-fucosyllactose, 3-fucosyllactose, and difucosyllactose by various probiotic and pathogenic bacteria.
Human milk glycomics and gut microbial genomics in infant feces show a correlation between human milk oligosaccharides and gut microbiota: a proof-of-concept study.
Effects of prebiotics on sepsis, necrotizing enterocolitis, mortality, feeding intolerance, time to full enteral feeding, length of hospital stay, and stool frequency in preterm infants: a meta-analysis.
Similar to those who are breastfed, infants fed a formula containing 2′-fucosyllactose have lower inflammatory cytokines in a randomized controlled trial.
Term infant formula supplemented with milk-derived oligosaccharides shifts the gut microbiota closer to that of human milk-fed infants and improves intestinal immune defense: a randomized controlled trial.
The effect of lactoferrin supplementation on death or major morbidity in very low birthweight infants (LIFT): a multicentre, double-blind, randomised controlled trial.
Human lysozyme expressed in the mammary gland of transgenic dairy goats can inhibit the growth of bacteria that cause mastitis and the cold-spoilage of milk.
Comparative proteomics of milk fat globule membrane (MFGM) proteome across species and lactation stages and the potentials of MFGM fractions in infant formula preparation.
Fecal microbiome and metabolome of infants fed bovine MFGM supplemented formula or standard formula with breast-fed infants as reference: a randomized controlled trial.
Recombinant human-milk bile-salt-stimulated lipase. Functional properties are retained in the absence of glycosylation and the unique proline-rich repeats.
00459-8&id=gr1.jpg){kind=link}
00459-8&id=gr2.jpg){kind=link}
00459-8&id=gr3.jpg){kind=link}
00459-8&id=gr4.jpg){kind=link}