Human Milk Oligosaccharides, Important Milk Bioactives for Child Health: A Perspective

19 min read
Hanne L. P. Tytgat
Sean Austin
Dominik Grathwohl
Dr. Norbert Sprenger

Human milk contains all nutritive and bioactive compounds to give infants the best possible start in life. Human milk bioactives cover a broad range of components, including immune cells, antimicrobial proteins, microbes, and human milk oligosaccharides (HMOs). Over the last decade, HMOs have gained special attention as their industrial production has allowed the study of their structure-function relation in reductionist experimental setups. This has shed light on how HMOs steer microbiome and immune system development in early life but also how HMOs affect infant health (e.g., antibiotic use, respiratory tract infections). We are on the verge of a new era where we can examine human milk as a complex biological system. This allows not only study of the mode of action and causality of individual human milk components but also investigation of synergistic effects that might exist between different bioactives. This new wave in human milk research is largely fueled by significant advances in analytical tools in the field of systems biology and network analysis. It will be exciting to explore how human milk composition is affected by different factors, how different human milk compounds work together, and how this influences healthy infant development.

Human Milk Supports Early Life Development

Human milk is the optimal source of nutrition to the infant in early life with lactose, proteins, and fat providing the energy and building blocks to support infant growth and development. It is also the source of a myriad of bioactive components that are not digested, but support and impact the development of the infant’s immune system and microbiome. These components come in many forms: from immune cells to sphingolipids, polar lipids, antimicrobial proteins, microbes, and human milk oligosaccharides (HMOs) to name just a few (Fig. 1). HMOs are nondigestible elongations of lactose that form the third most abundant solid component of human milk. Over 160 HMO structures have been described, the bulk of which are unique to human milk and only to a minor extent found in farmed animal milk. They can be grouped into three major classes based on their structure: nonfucosylated neutral, fucosylated, and sialylated HMOs [1].

Human milk and its compounds have been widely studied in observational cohorts, but the most profound insights stem from evaluation of the role of single or combinations of bioactives in randomized controlled clinical trials (RCTs). Advances in biotechnology have made the production of several of the most abundant HMOs of human milk possible on a large scale, allowing the clinical evaluation of their effect on several aspects of an infant’s development like the microbiome, immune system, and cognition. These studies, backed up with in vitro assays, led to the elucidation of causal structure-function relationships.

Apart from these reductionist approaches, it is important to investigate human milk as a system, as different compounds might work together synergistically. Furthermore, its composition needs to be evaluated in the context of maternal health and lifestyle, physiological status, maternal genetics, lactation time, and geography. These are only a few of the factors that influence the presence and concentration of key bioactive compounds in human milk with presumed effects on infant health (Fig. 1). It is therefore important to look at human milk as a complex biological fluid, to comprehend how its different bioactives contribute to infant growth and development.

Below we zoom in on how the field is developing toward a more systematic evaluation of HMOs in context with other bioactives in the complex biofluid that is human milk.

HMOs Are a Key Component of the Diet-Microbiome-Health Triad in Early Life

An age-adapted microbiome is increasingly recognized as a key driver of immune, cognitive, and temperament development as well as (future) metabolic health. Microbiome dysbiosis, induced through C-section birth and/or antibiotic use, provides important insights into the link between the microbiome in early life and the risk of developing respiratory infections, allergies, and obesity as well as its influence on cognitive development [2–5]. The exposome, defined as the sum of environmental (i.e., nongenetic) exposures, has a major influence on the early life microbiome development and maturation with feeding mode, i.e., breast milk versus formula milk, being a major contributor [5, 6].

HMOs have been recognized as key components impacting the gut microbiome [7]. HMOs were found to vary in amounts and profile over the course of lactation, due to maternal physiological status and genetic polymorphisms. Indeed, polymorphisms of the fucosyltransferase 2 (FUT2) and FUT3 enzymes in the lactating mother define respectively her secretor status and Lewis status, which affect the presence and concentration of key HMOs like 2′-fucosyllactose (2′-FL) and 3-FL. Many studies have focused on studying HMO biology in breastfed infants, revealing associations between individual HMOs in mother’s milk and the infant’s microbiome, growth, and health, as recently reviewed [1].

Besides being a driver in microbiome maturation, association studies have indicated that HMOs also play key roles in affecting infant health. Observational and basic research efforts have proposed effects of HMOs on various pathologies such as diarrhea and respiratory tract infections [8, 9] but also how HMOs may modulate the risk of allergy and how (mainly sialylated) HMOs may impact neurocognition [1]. Ultimately, confirmation and support for hypotheses formulated on the biological role and specific action of individual HMOs require carefully designed, randomized, controlled, intervention trials (RCTs).

Several RCTs have been conducted with the primary aim to evaluate the safety and tolerance of infant formulas containing one, two, and even combinations of five individually manufactured HMOs. These trials confirmed that HMOcontaining term infant formula was safe, was very well tolerated, and allowed for age-appropriate growth [10–12]. These trials also help to investigate the link between HMOs, microbiome maturation, and health outcomes, further enhancing our knowledge of HMOs and human milk bioactives at large as key players in driving infant health. Indeed, Alliet et al. [10] reported that the gut microbiota of HMO formula–fed infants tracked toward that observed in breastfed infants, including a slight bifidogenic effect and a reduction of the opportunistic pathogen Clostridiodes difficile. In an RCT that evaluated the 2 HMOs, 2′-FL and LNnT, Puccio et al. [12] reported that infants in the HMO formula arm had a reduced risk of lower respiratory tract infections and reduced need for antibiotics and antipyretics.

In the RCTs that investigated two slightly different blends of five HMOs, Lasekan et al. [13] reported that HMO formula-fed infants needed doctor consultations less frequently compared to control formula-fed infants. Bosheva et al. [11] reported bifidogenic effects affecting growth and bacterial metabolic activity resulting in lower stool pH and higher relative acetate and lactate, with concomitant increase in sIgA and decreased alpha-1 antitrypsin indicating a favorable action on the gut barrier function and integrity.

The potential mechanisms, both microbiome dependent and independent, underlying the effects seen in observational and interventional studies, have to be further investigated in molecular assays (e.g., receptors or cell-based assays). Extra care is warranted when using and interpreting results from such assays to make sure they are appropriately controlled for artifacts. Data indicate that HMOs may affect several physiological mechanisms, either directly or indirectly: through modulation of specific microbes from the gut microbiome, influencing the epithelial barrier function or interaction with immune cells. Many layers and mechanisms are contributing to establish physiological functions.

Redundancy between some of the HMO species may represent a “safety net” and be part of the reason why human milk is so rich in diverse HMOs. As illustrated in Figure 2a, combinations of HMOs can activate a specific receptor, which can also be activated by a microbial metabolite boosted through yet another HMO [14]. This is an example illustrating that broader network and classification analyses are warranted to understand if and how different milk components depend on each other and act together at the molecular level [15].

Similarly, different HMOs may exert synergistic activity, boosting effects, for example, via specific Bifidobacterium strains that produce metabolites such as the short-chain fatty acid acetate that in turn exert several different physiological functions (Fig. 2b). It is likely that several bioactives, beyond HMOs, could act in a synergistic manner, either by targeting similar pathways or by triggering different pathways converging to a physiological effect. Unraveling such mechanisms can further enhance our understanding of how human milk steers infant development.

Studying HMOs within the Complex Matrix of Human Milk

Most studies so far have looked at HMOs in a reductionist manner, evaluating the effect of single compounds on different biological parameters. Given that some of these studies showcase the potential synergistic effects of HMOs (Fig. 2), it is important to look at HMOs, and by extension “all” human milk bioactives, in the context of the complex biological matrix, human milk. Interactions, synergisms, and even antagonisms might exist between different compounds and might influence physiological outcomes.

So far, very few studies have assessed associations between human milk bioactives. A recent study on a population from Malawi reported several dispersed correlations between HMOs, groups of HMOs, and bioactive proteins like lactalbumin, lactoferrin, lysozyme, antitrypsin, IgA, and osteopontin [1, 16, 17]. The investigators only tested associations of single components with growth and developmental outcomes. It would have been of interest to integrate multiple milk bioactives and then test multivariate associations with infant outcomes. Moossavi et al. [18] tried an integrated statistical approach to identify associations between different milk component groups like HMOs, fatty acids, and milk microbiota in the Canadian CHILD study and reported that these were stronger between milk components within each group than correlations between components from different groups.

An example of a data-driven approach to visualize the correlations among nutritive and non-nutritive components in the milk of 156 mothers from Germany showed moderate interdependency between measured single fatty acids and phospholipids, as well as between measured immune factors, although there was no correlation found with respect to child allergy status [19–21]. All these studies, which are limited by the available analytes, samples, and study design, are only small steps toward considering human milk as a biological system and not just a collection of single independent nutrients and bioactives. Such an approach will help us to understand better the “mother–milk–infant triad” [22, 23].

Human milk composition is influenced by a myriad of factors that are part of the key to understanding its biological function. Numerous studies, including systematic reviews, have presented data on the temporal changes of HMOs and other milk components during lactation in both term and preterm settings [1, 16, 17]. The majority of the studies sought to understand the concentrations and the dynamics of the milk components over time and occasionally how factors like maternal age, body mass index, socioeconomic status, and mode of delivery could be associated with these changes [1, 19, 24] (Fig. 1). In particular for HMOs, reports on their concentrations from different geographies are available, but to our knowledge no systematic review or meta-analysis is available to compare and discuss them [19–21].

The strong association of FUT2 and FUT3 genetic variants with concentrations of individual HMOs and groups of HMOs [22, 23] has led researchers to incorporate secretor status, Lewis status, and milk groups into their analyses when describing milk composition. The overall HMO content and profiles consistently differ as a result of maternal genetics (genetically or phenotypically defined), substrate availability, and glycotransferase competition [22, 23, 25]. Such interdependencies likely extend to other compound classes, underscoring the need to investigate and present milk composition as a whole system considering all available components analyzed.

Studies of the co-occurrence of milk components grouped by maternal genetics and over time should be extended and described in different contexts like maternal lifestyle, dietary habits and physiological/health status, pre-pregnancy body mass index, geographies, and ancestry. An integrated approach to describe milk composition as a whole and over time in different settings will enable us to (1) better understand maternal factors as well as changing nutritional needs of infants and the role of bioactive molecules during early development; (2) design maternal and infant interventions for infant health and development; and (3) address maternal needs during breastfeeding and optimize infant nutrition.

Additionally, we can look at biology-driven approaches inspired by the functional structures present in human milk, like the milk fat globule and exosomes, and start employing analytical methods to better understand which molecules are present there, how they change over time, and their association with maternal and infant factors [26]. For example, which are the core constituents of these structures and which of them increase or decrease based on time of lactation, maternal and infant health, and maternal diet? First reports of milk miRNA, a major component of exosomes, showed that indeed there is a core human milk miRNAome consisting of about ∼600 miRNAs, which are expressed along the first 4 months of lactation [27]. Integrating the composition of other cargo bioactives, like proteins and lipids, encapsulated into these exosomes would be the next step for informing the subsequent functional role of these macrovesicles.

When embarking on endeavors to understand human milk, it is also important to consider several technical aspects. Milk sampling and storage are known to impact the concentration of macronutrients and micronutrients; therefore, standardized protocols for collection and storage have been proposed [28–30]. Human milk analyzers using mid- and near-infrared spectroscopy are now widely used for macronutrient measurement and can be applied for different settings [31, 32]. Care must be taken, however, when using data from such instruments as the results do not always match those of reference methods [31]. Ongoing advances in metabolomics could further allow comprehensive qualitative analysis of bioactives, but these should be complemented with classic, targeted, absolute quantitative methods [33, 34]. Historical data and samples can be used to inform new research questions and hypotheses, but ultimately future studies designed to test specific hypotheses are necessary to advance the field from the current descriptive data-generating approach.

The efforts to explore a system, human milk, in all its complexity (i.e., components and external influencing factors) are examples of systems biology, a relatively novel way to look at biological systems as a whole using integrated approaches of analyzing and modeling. Metabolic networks can be used to discover synergies between breast milk components and epidemiological characteristics of mothers and infants, as showcased by Michel et al. [35]. There is a need for further application of networks to HMO data to better understand the models and their data.

One issue common to large datasets is missing values. Datasets including breast milk composition and epidemiological characteristics may encompass several hundreds of mother–infant dyads, but these datasets are rarely complete. It is crucial to consider the impact of missing values and how those are handled during data analysis. Of note, a network analysis can be run already from n = 100 mother–infant dyads. When mining many variables in large datasets, the risk of chance findings should be considered and validated in separate datasets or with complementary methods.

In summary, several studies describe the composition of bioactives, like HMOs, over time and their dependence on maternal and infant factors; very few studies have employed systems biology to understand human milk as whole. Human milk and HMO research needs to shift toward a more comprehensive approach, utilizing clinical study settings with standardized collection protocols, along with advanced wet and dry lab analytical methods, to understand the biology and “mother–milk–infant triad” and advance maternal lactation support for infant nutrition.


Human milk provides all necessary nutrients and molecules to give infants the best possible start in life. We know that human milk contains a myriad of bioactive compounds, but further research is necessary to determine the impact of these bioactives and their combinations on infant development and health. The evaluation of human milk as a complex biological fluid will further advance our understanding of how human milk supports the healthy development of infants in early life. Through the combination of both reductionist and more systemic network analysis efforts, answers can be provided to better understand structure–function aspects of human milk components and isolate the relevance, causality, synergy, and mode of action of the different molecules. This will lead to a deeper understanding of how human milk contributes to infant development and health, which will aid in ameliorating both infant and maternal health.


  1. Sprenger N, Tytgat HL, Binia A, et al. Biology of human milk oligosaccharides: from basic science to clinical evidence. J Hum Nutr Diet. 2022;35(2):280–99. 
  2. Korpela K, de Vos WM. Early life colonization of the human gut: microbes matter everywhere. Curr Opin Microbiol. 2018;44:70–8. 
  3. Vissing NH, Chawes BL, Rasmussen MA, et al. Epidemiology and risk factors of infection in early childhood. Pediatrics. 2018;141(6):e20170933. 
  4. Slykerman RF, Coomarasamy C, Wickens K, et al. Exposure to antibiotics in the first 24 months of life and neurocognitive outcomes at 11 years of age. Psychopharmacology (Berl). 2019;236(5):1573–82.
  5. Shao Y, Forster SC, Tsaliki E, et al. Stunted microbiota and opportunistic pathogen colonization in caesarean-section birth. Nature. 2019;574(7776):117–21.
  6. Baumann-Dudenhoeffer AM, D’Souza AW, Tarr PI, et al. Infant diet and maternal gestational weight gain predict early metabolic maturation of gut microbiomes. Nat Med. 2018;24(12):1822–29. 
  7. Sanchez C, Fente C, Regal P, et al. Human Milk Oligosaccharides (HMOs) and infant microbiota: a scoping review. Foods. 2021;10(6):1429.
  8. Newburg DS, Ruiz-Palacios GM, Altaye M, et al. Innate protection conferred by fucosylated oligosaccharides of human milk against diarrhea in breastfed infants. Glycobiology. 2004;14(3):253– 63. 
  9. Binia A, Siegwald L, Sultana S, et al. The influence of FUT2 and FUT3 polymorphisms and nasopharyngeal microbiome on respiratory infections in breastfed Bangladeshi infants from the microbiota and health study. mSphere. 2021;6(6):e0068621. 
  10. Alliet P, Vandenplas Y, Roggero P, et al. Safety and efficacy of a probiotic-containing infant formula supplemented with 2′-fucosyllactose: a double-blind randomized controlled trial. Nutr J. 2022;21(1):11. 
  11. Bosheva M, Tokodi I, Krasnow A, et al. Infant formula with a specific blend of five human milk oligosaccharides drives the gut microbiota development and improves gut maturation markers: a randomized controlled trial. Front Nutr. 2022;9:920362. 
  12. Puccio G, Alliet P, Cajozzo C, et al. Effects of infant formula with human milk oligosaccharides on growth and morbidity: a randomized multicenter trial. J Pediatr Gastroenterol Nutr. 2017;64(4):624–31. 
  13. Lasekan J, Choe Y, Dvoretskiy S, et al. Growth and gastrointestinal tolerance in healthy term infants fed milk-based infant formula supplemented with five human milk oligosaccharides (HMOs): a randomized multicenter trial. Nutrients. 2022;14(13):2625. 
  14.  Foata F, Sprenger N, Rochat F, et al. Activation of the G-protein coupled receptor GPR35 by human milk oligosaccharides through different pathways. Sci Rep. 2020;10(1):16117. 
  15. Miliku K, Robertson B, Sharma AK, et al. Human milk oligosaccharide profiles and food sensitization among infants in the CHILD Study. Eur J Allergy Clin Immunol. 2018;73(10):2070–73. 
  16. Gidrewicz DA, Fenton TR. A systematic review and meta-analysis of the nutrient content of preterm and term breast milk. BMC Pediatr. 2014;14:216. 
  17. Thum C, Wall CR, Weiss GA, et al. Changes in HMO concentrations throughout lactation: influencing factors, health effects and opportunities. Nutrients. 2021;13(7):2272. 
  18. Moossavi S, Atakora F, Miliku K, et al. Integrated analysis of human milk microbiota with oligosaccharides and fatty acids in the CHILD cohort. Front Nutr. 2019;6:58. 
  19. Samuel TM, Binia A, de Castro CA, et al. Impact of maternal characteristics on human milk oligosaccharide composition over the first 4 months of lactation in a cohort of healthy European mothers. Sci Rep. 2019;9(1):11767. 
  20.  McGuire MK, Meehan CL, McGuire MA, et al. What’s normal? Oligosaccharide concentrations and profiles in milk produced by healthy women vary geographically. Am J Clin Nutr. 2017;105(5):1086–100. 
  21. Zhou Y, Sun H, Li K, et al. Dynamic changes in human milk oligosaccharides in Chinese population: a systematic review and meta-analysis. Nutrients. 2021;13(9):2912. 
  22. Lefebvre G, Shevlyakova M, Charpagne A, et al. Time of lactation and maternal fucosyltransferase genetic polymorphisms determine the variability in human milk oligosaccharides. Front Nutr. 2020;7:574459. 
  23. Bode L. Human milk oligosaccharides: every baby needs a sugar mama. Glycobiology. 2012;22(9):1147–62. 
  24. Azad MB, Robertson B, Atakora F, et al. Human milk oligosaccharide concentrations are associated with multiple fixed and modifiable maternal characteristics, environmental factors, and feeding practices. J Nutr. 2018;148(11):1733–42.
  25. Siziba LP, Mank M, Stahl B, et al. Human milk oligosaccharide profiles over 12 months of lactation: The Ulm SPATZ Health Study. Nutrients. 2021;13(6):1973. 
  26. Ballard O, Morrow AL. Human milk composition: nutrients and bioactive factors. Pediatr Clin North Am. 2013;60(1):49–74.
  27.  Raymond F, Lefebvre G, Texari L, et al. Longitudinal human milk miRNA composition over the first 3 mo of lactation in a cohort of healthy mothers delivering term infants. J Nutr. 2022;152(1):94–106. 
  28. Samuel TM, Zhou Q, Giuffrida F, et al. Nutritional and non-nutritional composition of human milk is modulated by maternal, infant, and methodological factors. Front Nutr. 2020;7:576133. 
  29. Becker GE, Smith HA, Cooney F. Methods of milk expression for lactating women. Cochrane Database Syst Rev. 2015;2:CD006170. 
  30. Leghi GE, Middleton PF, Netting MJ, et al. A systematic review of collection and analysis of human milk for macronutrient composition. J Nutr. 2020;150(6):1652–70. 
  31. Giuffrida F, Austin S, Cuany D, et al. Comparison of macronutrient content in human milk measured by mid-infrared human milk analyzer and reference methods. J Perinatol. 2019;39(3):497– 503. 
  32. Perrin MT, Festival J, Starks S, et al. Accuracy and reliability of infrared analyzers for measuring human milk macronutrients in a milk bank setting. Curr Dev Nutr. 2019;3(11):nzz116. 
  33. Hampel D, Shahab-Ferdows S, Hossain M, et al. Validation and application of Biocrates AbsoluteIDQ® p180 targeted metabolomics kit using human milk. Nutrients. 2019;11(8):1733. 
  34. Sumayao R, Beningo SJ, Jean GA, et al. Breastmilk metabolomics: bridging the gap between maternal nutrition and infant health outcomes. KIMIKA. 2017;28(2):1–12. 
  35. Michel L, Shevlyakova M, Ni Cleirigh E, et al. Novel approach to visualize the inter-dependencies between maternal sensitization, breast milk immune components and human milk oligosaccharides in the LIFE Child cohort. PLoS One. 2020;15(4):e0230472.