The tools to understand lactation are from genomics to physiology. Whole genome sequencing provides the basic knowledge set. The challenge is to identify and annotate genes associated with lactation. The goals for lactation genomics were propelled by a diverse group of scientists participating with the International Milk Genomics Consortium (5). The diversity of lactation across mammalia is an important asset, so the goal of the IMGC was to assemble a variety of entire genomes from marsupials to humans (6). Comparing these genomes formed the basis for interpreting the evolution of mammalian lactation (7), the expression of genes during lactation (8), the sequences (9) and digestibility of proteins (10), the sequences of peptides (11) and their formation (12), the biosynthetic pathways of glycans (13), the variation among women due to genetic diversity (14), timing of lactation (15), diet (16) and the biosynthesis of lipids (17).

Milk is a challenge for analytical chemistry. From small molecules to entire cells, milk is a cornucopia of biomolecules varying in size and concentration by orders of magnitude. All the essential nutrients are in milk, each present within a matrix that enhances their bioavailability and controls their chemical reactivity. These matrices that enhance bioavailability impede the characterization of the molecules analytically. Every class of biopolymer is present in milk, all the substrates and intermediates in their synthesis. Milk oligosaccharides were typical. Entire methodological platforms had to be developed for this one biopolymer class alone (18). Oligosaccharide method development required innovative approaches to initial separation, liquid chromatography on novel stationary matrices, mass spectrometry techniques including highly sensitive time of flight and triple quadropole mass spectrometry. The construction of structurally annotated databases of mass spectra was necessary to automate high throughput (19).

The scientific investigations to discover biological actions of milk require that the components be available as purified research materials in quantities and purities maintained in their native conformations for the multiple assays by which investigators address their hypotheses. Also, human milk is a rare and valuable material and simply accessing milk as research material is and should be a process of regulatory, safety and ethical formalities. Combinations of traditional separation technologies and bioguided separations were needed (20). Strategies such as physical separation of milk components by size achieved enrichment in oligosaccharides but retained contamination by peptides and lactose (21). Peptides are an important biological resource in milk but vary widely in abundances and structures depending on the stages of lactation, treatment of milk etc. (22).

Glycans are abundant across the tree of life and the most abundant biopolymer in the biosphere (25). Despite their importance they are not sequence encoded but products of enzymatic metabolism. The enzyme specificity to produce glycan structures limits the number of biological structures that are found relative to the enormous number of structures that are mathematically possible. This difference between biologically feasible and mathematically possible has led to the concept of bio-defined analytics (C. Lebrilla, 2000, unpublished). Combining biology with chemical analysis has guided analytical strategies to catalog the glycan structures present in milk and a variety of organisms (26). Structures of glycans include monosaccharide composition, branching, the stereospecific linkages of those sugars all leading to multiple isomers even for a single net atomic mass. Glycan structures are both free and bonded to proteins, peptides or lipids again by enzymatic synthesis. Every glycan, in each sample, must be explicitly analyzed to be identified (14).

The oligosaccharides of human milk have been attractive to scientists because they are free, abundant (1-2% w/v) and yet indigestible by the neonate. They are perplexing to annotate: why would mothers “dissolve themselves” to produce these biopolymers in such abundance? The scientific challenges posed by this apparent paradox propelled laboratories to pursue the analytical platforms to identify and annotate them (27).

The structures of milk oligosaccharides have been selected, in part, for an unusual biological value: NOT to be consumed by infants. Research on oligosaccharides in human milk has established as one function, that they support the growth of specific bacteria notably strains of the genus Bifidobacterium (28). While the mechanisms and extent of microbial diversity in breastfed infants are still being actively documented, the basic observation that bifidobacterial species dominate the microbiota of breastfed infants around the world compared with formula-fed infants has been well-established (29). How an intestinal microbial ecosystem maintains a dominant and consistent bacterial population in the face of repeated and diverse inoculations with environmental microorganisms has been largely speculative until recently. The idea launched by Gyorgi that oligosaccharides were a Bifidus factor (30) was unfortunately insufficiently specific. Oligosaccharides do not stimulate the growth of the entire genus of Bifidobacterium in general. Bifidobacterium represent a broad genera of bacteria whose members occupy a wide range of ecological niches. Though first identified microscopicslly by Tissier in the 19th century in breast fed infants only recently has research recognized the unusual specificity of the strains of Bifidobacterium that dominate the intestinal microbiome of breast fed infants (31–33). Intensive studies revealed the remarkable interaction between the stereospecific linkages of milk oligosaccharides and the genetic repertoire of glycosidases and solute binding proteins that provide these bacteria a distinct competitive growth advantage within the intestine of the breast fed infant (34).

The colonization of the infant by microorganisms begins at birth (35). The consensus of microbiome research argues that the infant gut is ostensibly sterile at birth and those organisms that may have arrived into the amniotic compartment prior to delivery are not competitive once the “flood” of exogenous microorganisms (bacteria, yeast, fungi, viruses) that accompany a normal human birth. These initial inocula are the first of a continuous wave of inoculations of the infant from the environment (36). The mode of delivery, vaginal or by C-section has been noted to alter the gut microbiota of term infants in early life (37), however, these observations are mainly of infants within a restricted microbial environment, the modern hospital delivery room. Infants delivered vaginally acquire bacterial communities resembling those of maternal vagina and fecal microbiomes, while C section babies initially reflect a microbiota resembling that of maternal skin. Infants delivered vaginally exhibited higher abundances of Bacteroidaceae and lower abundances of Enterococcaceae, Pasteurellaceae, Carnobacteriaceae, and Gemellaceae compared to C section delivered infants (38). Knowing that each infant is inoculated with a diverse array of organisms, a goal was to understand the role of that environment and milk components simultaneously in guiding the distinct microbiological community in the breast fed infant. Which microorganisms utilize and grow on specific components of milk (39)? Many components from milk, in isolation, can support microbial growth. Thus, enabling technologies were needed: isolating potential growth substrates in pure form from milk and media for bacterial culture assays that include all of the nutrient requirements for growth, but lack a carbon fuel source. Into these media can then be added the components of milk that are expected to arrive at different sections of the intestine (39). The complex oligosaccharides from milk were isolated to assess bacterial growth on those undigestible components of milk that arrive at the lower intestine. Surprisingly, initial growth experiments did not observe significant growth of bacteria when human milk oligosaccharides were the sole source of carbon in the otherwise supportive medium (40). Among gut-related bacteria tested (including Lactobacillus, Clostridium, Eubacterium, E. coli, Veillonella, Enterococcus isolates) only Bifidobacterium and Bacteriodes species grew to high cell densities yet, growth was strain specific (41). Robust growth on HMO was found just in a select group of B. bifidum and B. longum subsp. infantis (B. infantis) strains. In these same growth conditions even isolates of B. longum subsp. longum and B. breve showed poor growth and strains of B. adolescentis, and B. animales were unable to grow on HMO (41).

Any ecosystem is driven by accessible food. The lower intestine of the breast fed infant is supplied by those components of milk that are not digested nor absorbed by the infant in the upper intestine. Thus, those bacteria capable of accessing oligosaccharides are provided a competitive advantage by the infant's mother's milk. Nonetheless, only the combination of microorganisms growing on the oligosaccharides coded by lactation genes from each infant's mother that confer a selective advantage to infant success would be rewarded through evolution. The outcomes of that Darwinian engine, pathogen protection to immune education are continuing to emerge as novel mechanisms of Bifidobacterium dominated microbiome actions (42).

One defining set of traits for colonic bacteria is their ability to degrade biopolymers and access the monomeric sugars, amino acids etc., in that environment. How they do that is important. Most intestinal bacteria secrete extracellular glycosidase enzymes into the luminal environment and these enzymes catalyze the hydrolysis of complex glycans and liberate free sugars extracellularly. Free sugars are taken up by bacteria and metabolized. Select strains of bifidobacteria use extracellular lacto-N-biosidase activity to break down oligosaccharides (43). Some bacterial strains, notably B. infantis, pursue a different strategy of transporting oligomeric structures into the interior of the cell and breakdown reactions occur internally. This internal feeding strategy confers an advantage to the host by blocking the liberation of simple sugars into the lumen that other organisms can utilize. Cross feeding liberated sugars to other organisms is a known mechanism to promote the growth of undesirable, opportunistic pathogens (44).

The discovery that growth of bacteria on milk oligosaccharides was a strain specific, gene driven process and that B. infantis ATCC15697 was uniquely capable phenotypically, prompted the goal to sequence its genome and begin the process of annotating its unique capabilities. One of the joys of being a scientist is those occasions when you are witness to the sheer elegance of biology. The genetic repertoire of B. infantis, was one of those rare moments in which scientific discovery revealed that elegance (31). This specific strain provides the field of microbiome research with insights into the traits associated with capabilities to thrive within the anaerobic intestine including genes providing the strain its phenotype (31). Breast fed infants that are exposed to such HMO consuming strains are colonized by them and in turn achieve direct and indirect benefits. Those benefits even include the protection from the horizontal transfer of virulence and antibiotic resistance traits (45). These benefits are consistent with the concept that the oligosaccharides produced by the mammary gland and the emergence of oligosaccharide consumption gene clusters in specific strains of bifidobacterial strains are an example of symbiotic co-evolution.

The principle of nourishment as the center of cross-kingdom partnerships is not unique to lactation. Glycan based nourishment appears to be at the center of most cross-kingdom symbioses from plants feeding pollinating insects with sugar rich nectar (46) to roots feeding nitrogen fixing bacteria (47). This same strategy emerging from evolution of milk feeding a metabolically distinct and mutually beneficial bacterial population (mutualism) in infants is another example. The challenge is what do we learn by understanding it?

Evidence from epidemiology, mechanistic insights and increasingly prospective interventions shows that the mutualism between human breast milk and the B. infantis commensal is important, yet fragile. The importance was first suggested by premature infants. Infants born premature, by Cesarian section, are placed in an incubator. At that point the immediate hospital environment serves as the inoculating reservoir of seeding microorganisms. In such an environment, the explicit steps taken to prevent cross-patient pathogen transfer, (scrupulous hygiene, sanitation, etc.) have the unintended consequence of preventing the transfer of commensal organisms as well. The first indications of the outcomes of that environment emerged in studies comparing the explicit inoculation of candidate organisms. Studies used in-vivo administration of B. infantis to premature infants fed either formula or breast milk. Breast milk-fed infants, when supplemented with B. infantis saw increases in fecal bifidobacteria and decreases in γ-Proteobacteria compared with a formula-fed group (48).

Following on those initial studies, B. infantis, used clinically, has already been demonstrated to significantly impact the development of inflammation (49), autoimmunity (50) and necrotizing enterocolitis and mortality of premature infants (51). Thus, understanding how mothers are shaping the protective milk-oriented microbiota (MOM) of their infants through breast milk is an urgent model for guiding microbial communities at all ages.