During a process called adipogenesis, progenitor cells develop into preadipocytes and subsequently mature, lipid-laden adipocytes (27). Adipocyte progenitor cells are committed to becoming mature adipocytes in fetal and early postnatal life (28, 29). Leptin and adiponectin can stimulate the differentiation from preadipocytes into adipocytes in vitro (30, 31). It should however be noted that the literature reports contrasting findings for breast milk hormones and adipokines in relation to adiposity in vivo, which is discussed in a recent systemic review by Mazzocchi et al. (4). Further studies are needed to clarify the roles of bioactive components in breast milk in relation to body composition. Other breast milk factors such as lactoferrin (32) and short-chain fatty acids (SCFA) (33, 34) have also been found to increase adipogenesis in vitro. SCFA are microbiota-derived metabolites that are most likely produced by the maternal gut microbiota and enter the milk via the circulation (35). In addition, milk oligosaccharides are fermented in the colon of the neonate generating SCFA (36, 37). SCFA can stimulate adipogenesis through G protein-coupled receptors (GPCR) signaling (33) and inhibition of histone deacetylase activity (34). Thus, as epigenetic regulators SCFA may have the potential to program adiposity. On the other hand, inflammatory cytokines such a TNF-α reportedly inhibit adipogenesis (38, 39), highlighting the potential importance of anti-inflammatory compounds in human milk. While there are no studies today on the direct influence of human milk on adipose tissue development, we can postulate that the nutrient content and anti/pro-inflammatory profile of human milk will influence adipocyte ontogeny. This deserves to be addressed in future research.

Excess nutrients are stored in white adipocytes as lipid droplets in response to insulin (40). Studies in mice have identified that once adipocytes increase in size in response to excess nutrient load during the early post-weaning period, a subsequent expansion of progenitors occurs due to increased systemic IGF-1 levels (41). If these cells encounter another dietary challenge in adulthood they increase in numbers, resulting in higher adipose tissue mass (41). Protein restriction in mice at weaning resulted in reduced adipocyte proliferation and growth restriction. These young mice increased their food intake in order to adapt, and this continued into adulthood (42). In culture, adiponectin has been shown to promote proliferation and increase lipid content of adipocytes (31). These studies show the importance of the diet for lipid accumulation and adipocyte proliferation, suggesting a significant role for the growth-adapted energy and nutrient content of human milk for adipose tissue expansion in early life.

Neonates are born with beige adipose tissue-dominated fat depots. The energy content of human milk has the potential to influence adipose tissue beiging, as it has been shown that restricting calorie intake in young mice reduces the expression of brown adipose tissue (BAT) markers such as uncoupling protein 1 (UCP-1) in WAT at weaning (43). Another study has demonstrated that a high-fat diet in young mice that were exposed to a low-protein diet prenatally prevented the differentiation of precursors into beige adipocytes (44). Leptin and adiponectin promoted beiging of WAT in mice (45, 46) and lactoferrin upregulated UCP-1 expression in brown adipocytes in culture (47). The lipid composition of human milk is key for beige adipose tissue development. A recent study in mice demonstrated that breast milk-specific alkylglycerols sustain beige adipocytes through adipose tissue macrophages. Macrophages metabolize the alkylglycerols into platelet-activating factor (PAF), which leads to IL-6 secretion. Subsequently IL-6/STAT3 signaling in adipocytes triggers beige adipose tissue development. Therefore, breast milk intake delays the reduction of beige fat in early life, and reduces fat accumulation (48). Importantly, these lipids are absent in formula or the adult diet and the replacement of beige adipose tissue by WAT is accelerated in obese children (49). In addition, human milk SCFA levels are inversely associated with infant weight gain and adiposity and thus might protect from excess weight gain in early life (50). Although the exact mechanisms remain to be determined, milk-derived SCFA may contribute to the regulation of adiposity via the activation of BAT increasing energy expenditure (51, 52). These studies suggest that human milk energy content and lipid composition are key for healthy adipose tissue development and energy expenditure.

The adipose tissue does not only comprise adipocytes as its stromal vascular fraction contains heterogeneous cell populations such as mesenchymal progenitor/stem cells, preadipocytes, endothelial cells, pericytes and immune cells. Data from adult experimental models highlight that these cells also play an important role in controlling adipocyte energy metabolism. Regulatory T cells (Treg) support adipose tissue function by keeping local and systemic inflammation in check and promoting insulin sensitivity (53). Both visceral and subcutaneous WAT have low fractions of CD4+ FoxP3+ Treg cells at birth. In the visceral WAT Tregs accumulate over time and account for over half of T cells in lean adult mice (54). Besides Treg, type 2 immunity is required in the adipose tissue for optimal energy metabolism as it leads to the activation of eosinophils and alternately activated M2-type macrophages. Together these cells act to promote browning, which can reduce adiposity (55, 56) as described above.

Human milk has the potential to assist in adipose tissue homeostasis by preventing chronic systemic inflammation, which is linked to preserving the gut mucosal barrier (discussed in more detail below). Reducing circulating lipopolysaccharide (LPS) concentrations lowers pro-inflammatory gene expression in visceral WAT, dampening local WAT inflammation and improving insulin sensitivity (57). As lipids do not only serve as an energy substrate, but also have an important role in cell signaling, the quality of lipids in human milk is of interest for the development and inflammatory status of the neonate. Long chain n-3 polyunsaturated fatty acids (PUFA) are generally considered protective against inflammation (58) and have been shown to decrease the infiltration and cytokine production of inflammatory macrophages in adult adipose tissue (59). Although the key role of PUFA for development in early life (e.g. for neural development) is widely recognized, its importance for growth and metabolic health in infancy remains unclear. The interaction of SCFA with adipose tissue is relevant for the prevention of metabolic disease and its associated inflammation (60). SCFA have anti-inflammatory properties and reduce the expression of inflammatory cytokines and chemokines as well as leukocyte infiltration in adipose tissue (61). SCFA however also induce leptin secretion, which can induce a proinflammatory cytokine profile (30, 60, 62). Lactoferrin has been demonstrated to decrease inflammatory markers in adipocytes (32). The relevance for this in early life remains unknown and requires further investigation. However, the examples above illustrate that mediators present in human milk, or generated in the infant’s gut from milk components, can support adipose tissue homeostasis.

The infant’s gut starts off permeable due to its immaturity (63, 64), allowing for increased intestinal absorption of intact proteins compared to the adult (65, 66). This may facilitate the intact absorption of bioactive factors present in breast milk. As touched upon above, from the adult situation it is known that defects in the integrity of the gut barrier can also result in systemic translocation of bacterial/gut-derived material leading to endotoxemia, which contributes to chronic low-grade inflammation and metabolic dysfunction (67–69). In healthy cases a structural barrier of epithelial cells proliferates in early life and restricts the entry of foreign (bacterial) material into the systemic circulation (63, 70). Studies in young infants demonstrate the major beneficial effect of human milk as compared to formula for the development of gut barrier integrity (64, 71). This is most probably related to breast milk factors required for epithelial cell maturation and function, including epidermal growth factor (EGF) (6, 72), IGF-1 (22), transforming growth factor (TGF)-β (73), lactoferrin (74), vaso-active intestinal peptide (VIP) (75), vitamin A (37, 76) and SCFA (36, 37). Analysis of epithelial cells obtained from stool samples at 3 months of age demonstrated differential expression of over 1,000 genes between breastfed and formula-fed infants (77). The importance of the gut barrier in early life for energy metabolism is highlighted in infants with chronic intestinal inflammation. These infants show poor nutrient absorption leading to macro- and micronutrient deficiencies and the use of energy for the inflammatory response, resulting in impaired growth and physical development (78).

A population of innate gut immune cells, innate lymphoid cell type 3 (ILC3), represents a potentially important target for control of energy metabolism by breastfeeding. These cells develop in utero under the influence of maternal vitamin A and aryl hydrocarbon receptor (AhR) ligands (79, 80). AhR ligands, originating from the maternal microbiota, are also detected in breast milk and stimulate the development of ILC3 in the intestine of the neonate (80, 81). Mouse models suggest that these cells could be key in the regulation of inflammation and energy metabolism after weaning. Through the secretion of IL-22, ILC3 control the expansion of inflammatory bacteria in the gut (82) and promote a strong epithelial barrier (57), both preventing chronic inflammation. IL-22 also enhances PYY levels, an anorexic gut hormone reducing food intake (57), and is likely involved in the control of lipid uptake. Studies show that IL-22 reduces lipid transport (82) resulting in lower body weight (83) while others describe that IL-22 enhances lipid uptake and export to the adipose tissue, and therefore increase body fat (84). IL-22 secretion by ILC3 is controlled by bacterial activation of Toll-like receptor-MyD88 signaling (46). VIP, secreted from enteric neurons upon food consumption, also appears important for regulation of IL-22-producing ILC3 in the gut. However, divergent effects were observed in two recent studies (83, 85). It may be of interest to investigate the role of VIP in regulating this pathway in early life, as human milk contains ~100pg/mL VIP (86). The studies described here show that various milk components can improve the intestinal barrier and prevent chronic systemic inflammation and thus support adipose tissue homeostasis.

Perhaps the clearest links between human milk, the gut and metabolic outcomes are related to the gut microbiota. Microbial signals such as microbe-associated molecular patterns (MAMPs) and bacterial metabolites such as SCFA modify local mucosal and systemic immune responses (36), as described above. In the adult, changes in the gut microbiota composition influence energy extraction (87), inflammation (68, 88), metabolic signaling (89) and browning of adipose tissue (90). Because the intestinal microbiota establishes as the adipose tissue and immune system develop, microbiota disruption during the early time window can impair host metabolism, body composition and healthy postnatal growth. Infants with undernutrition show persistent microbiota immaturity (91–93) which correlates with anthropometric growth measurements (93). On the other hand, Caesarean section (94–96), lack of breastfeeding (3) and the use of antibiotics (9–11) are associated with gut dysbiosis and being overweight in childhood. The impact of early life antibiotics on adiposity can remain after ceasing of antibiotic treatment, even though the microbiota recovers, highlighting that alterations in the critical window can have long-term effects (97, 98).

Breastfeeding status is the most significant factor associated with microbiota structure in early life (99–104). Breast milk bacteria contribute to the seeding of the infants gut microbiota, and this is dependent on breastfeeding exclusivity and duration (102). These bacteria most likely originate from the mother as well as other exogenous sources such as the home environment and infant mouth (102). Bacteria transferred through human milk include species such as Rothia, Veillonella and Bifidobacterium that can influence immune homeostasis (102). The microbiota of healthy, exclusively breastfed infants is dominated by Bifidobacterium, which is linked to health benefits such as improved gut barrier function (92). A recent cohort showed that the abundance of Enterococcus in the infant gut, seemingly due to the lower abundance in maternal breast milk, inversely correlated with infant body weight and fat as well as leptin levels (105). A recent mouse study identified bile acids as potent drivers of intestinal microbial maturation in early life (106), and potentially human milk bile acids may contribute to this effect. Human milk also provides the neonate with large amounts of antimicrobial factors such as lactoferrin, which can further shape the early microbiota (103). Of large interest to date are prebiotic factors such as human milk oligosaccharides (HMOs), which support the growth of beneficial bacteria and can alter colonization patterns in the neonate (103). HMOs are a group of complex sugars that are the third most abundant solid component of human milk, and though non-nutritive to the infant they have many properties besides their role as prebiotics. The influence of HMOs on growth is therefore most probably multi-factorial and beyond its prebiotic activity includes inhibition of pathogen adhesion and host invasion, local and systemic anti-inflammatory effects (107) and modification of the metabolism in the microbiota as well as the host (108). A study amongst Malawian mothers with undernourished infants has demonstrated that milk oligosaccharides are less abundant in breast milk of mothers with stunted infants. This study also showed a causal, microbiota-dependent relationship between sialylated milk oligosaccharides and growth promotion in malnourished animals (108). An elegant mouse study by Cowardin et al., demonstrated that sialylated milk oligosaccharides, but not the most abundant HMO 2’fucosyllactose (2’FL), decreased bone resorption, leading to an increase in bone volume and growth in a similar low resource setting (109). In a healthy population, human milk 2’FL was positively associated with growth in infancy and early childhood, whereas high concentrations of lacto-N-neo-tetraose (LNnT) were associated with reduced body fat and growth (110–112). Overall, HMO diversity was inversely associated with infant fat mass (110) and childhood growth (112). Maternal exercise induced 3’-siallyllactose in human milk, and this HMO improved metabolic health and protected from the harmful effects of high-fat diet feeding in adult mouse offspring (113). These recent studies all point out the importance of HMOs for healthy growth in early life.