Selected Human Milk Oligosaccharides Added to Infant Formulas for Term Infants

Abstract

The benefits of breastfeeding, such as reduced risk of gastrointestinal and respiratory tract infections, depend largely on the presence of bioactive compounds in breast milk, including human milk oligosaccharides (HMOs). The presence of HMOs represents one of the largest differences in composition between breast milk and infant formula. Currently, progress in biotechnology allows the production of selected HMOs such as 2'-fucosyllactose (2'-FL) and lacto-N-neotetraose (LNnT), which are increasingly being added to infant formulas to narrow the difference between breast milk and formula. It is important to differentiate HMOs naturally occurring in human breast milk from those biotechnologically produced, which, while identical to HMOs in breast milk, do not originate from breast milk. This chapter summarizes basic facts about HMOs, findings from observational studies assessing the relationship between specific HMOs and clinical effects, and evidence from randomized controlled trials with structures identical to HMOs in breast milk added to infant formulas. Overall, the findings from some recently published trials provide reassurance that infant formulas supplemented with selected structures identical to HMOs, specifically 2'-FL with/out LNnT, are safe and well tolerated, and may have favorable effects on some health outcomes and medication usage. Further studies are needed.

Introduction

The benefits of breastfeeding, such as reduced risk of gastrointestinal and respi-ratory tract infections [1], depend largely on the presence of bioactive compounds in breast milk, including human milk oligosaccharides (HMOs) [2]. Over the past few years, HMOs have become one of the hot topics of research in infant nutrition. It results from the development of analytical methods that allow better understanding of the structures and properties of HMOs. Furthermore, at least some HMOs can be produced in large quantities using modern biotechnological methods. Some of these are added to infant formulas to narrow the gap between breast milk and infant formula. This article summarizes basic facts on HMOs and findings from (1) observational studies assessing the relationship between specific HMOs and clinical effects, and (2) randomized controlled trials (RCTs) with selected HMOs added to infant formulas. For the latter, a search of PubMed and Cochrane library was performed from 2015 to April 2021.

Human Milk Oligosaccharides

A detailed discussion of HMOs is beyond the scope of this article but can be found elsewhere [3, 4]. In brief, HMOs are complex carbohydrates that include five monosaccharides: glucose, galactose, N-acetylglucosamine, fucose, and sialic acid. In terms of quantity, HMOs are the third largest solid component of breast milk (only lactose and fats are present in greater amounts). It is estimated that breast milk contains about 150-200 structurally distinct HMOs, although the exact number is unknown. Each mother produces a specific HMO set, which is genetically determined (similar to blood groups). HMOs have no nutritional value for the infant. They primarily act as prebiotics by promoting healthy gut microbiota composition and serving as a selective source of nutrition for some microorganisms, mainly Bifidobacterium infantis, Bifidobacterium bifidum, and Bifidobacterium breve, inhabiting the gastrointestinal tract. These microorganisms “repay” the host by modifying the microbiota and contributing to a reduction in the risk of diseases, especially gastrointestinal infections. Additionally, HMOs act as antimicrobial agents by preventing pathogen adhesion to epithelial cells. HMOs are also involved in immune function acting as intestinal epithelial cell modulators, by enhancing maturation of the intestinal mucosa and intestinal epithelial barrier function, and as immunomodulators directly or indirectly modulating the immune system.

Findings from Observational Studies on Specific HMOs and Clinical Effects

In addition to a number of in vitro studies documenting the effects of HMOs such as immunomodulatory activity against viruses (influenza, rotavirus, respiratory syncytial virus, norovirus) [5], observational human studies suggest an association between the levels of both total HMOs and specific HMOs in breast milk and clinical effects in infants. Among others, the studies analyzed the associations between HMOs and outcomes such as [3]:
•    Gastrointestinal infections - a lower risk of Campylobacter jejuni-induced diarrhea in children who received breast milk containing large amounts of 2'-fucosyllactose (2'-FL) and a lower risk of Calicivirus-induced diarrhea in children receiving milk with a high content of another fucosylated HMO, lacto-N-difucohexaose I, was shown [6]
•    Allergies - the results of studies are not conclusive [7-9]; however, preliminary data suggest a link between fucosylated HMOs (2'-FL) and a reduction in the risk of allergies, especially in infants born via caesarean section [10]
•    Necrotizing enterocolitis - a link between low levels of disialyl-lacto-N-te- traose in breast milk and the risk of necrotizing enterocolitis in premature infants was documented [11, 12]
•    Sepsis - high concentrations of fucosyl-disialyllacto-N-hexaose in breast milk were associated with a reduced risk of death in very low birth weight infants with sepsis [13]
•    Overweight/obesity - the results of a small study showed differences in the HMO composition of breast milk from mothers of breastfed infants with normal and excessive weight gain; in the latter case, a higher 2'-FL content was found [14, 15]
Association, however, is not causation. Observational studies are not sufficient for proving that any specific HMOs would alter health. Still, the findings are important, as they indicate directions for further research.

Structures Identical to HMOs Added to Infant Formulas

Until recently, obtaining HMOs was very difficult. However, biotechnological methods, including microbial fermentation using genetically engineered micro-organisms E. coli and yeast [16], allow the manufacture of selected HMOs, in-cluding 2'-FL and lacto-N-neotetraose (LNnT), in very large quantities [17]. Since 2016, 2'-FL and LNnT have also been added to some infant formulas. Two leading institutions providing opinions on food-related issues (European Food Safety Authority [EFSA] [18] and Food and Drug Administration [FDA]) [19], independently of each other, have confirmed the safety of 2'-FL and LNnT when added alone or in combination to infant, follow-on, and young child formula. Other HMOs are also available (Table 1) [20]. However, it is unlikely that all HMOs found in breast milk can be produced in the near future (or are even needed).

Terminology

It is important to differentiate HMOs naturally occurring in human breast milk from those biotechnologically produced, which, while identical to HMOs in breast milk, do not originate from breast milk. Those biotechnologically manufactured HMOs are sometimes called “synthetic HMDs” [21] or “human-identical milk oligosaccharides” [22] or “artificial HMDs” [23] or “HMD analogues” [24]. Another option is the term “structures identical to HMOs.” However, there is no consensus regarding the terminology, and further discussions are needed to ensure general agreement and scientific clarity in communication.

Findings from RCTs of Infant Formulas Supplemented with 2'-FL and LNnT

Table 2 summarizes the key results of RCTs published as full papers (as of April 2021) assessing the safety and efficacy of infant formulas supplemented with 2'- FL with or without LNnT.
 

2'-FL Supplementation

A 2015 RCT by Marriage et al. [25] evaluated the effects of 2'-FL-supplemented infant formula on weight gain per day from day of life 14 to 119 (primary endpoint) as well as tolerance and other anthropometric measures (secondary out-comes). Healthy infants born of a single pregnancy, with a birth weight >2,490 g, were eligible for the study. Infants meeting the inclusion criteria were randomly assigned to one of 3 groups in which they received:
(1) infant formula containing galactooligosaccharides, with an energy value of 64.3 kcal/100 mL;
(2)    the same formula containing additional 2'-FL in the amount of 0.2 g/L, or
(3)    the same formula containing additional 2'-FL in the amount of 1 g/L. The reference group consisted of infants who met the eligibility criteria but were ex-clusively breastfed. The intervention, exclusive feeding of formula or breast milk, lasted until 119 days of age. Baseline clinical and demographic characteristics did not differ significantly between the intervention groups. Similar increases in weight, body length, and head circumference were found in all study groups, with growth parameters in the formula-fed groups similar to those found in exclusively breastfed infants. 2'-FL-supplemented formula was well tolerated. At 42 days of life, relative absorption of 2'-FL in the experimental formula groups was similar to that seen in breastfed infants.

In another publication [26], some of the same authors presented the results of an evaluation of the profiles of proinflammatory cytokines (TNF-a, IL-1a, IL- 1p, IL-6) and anti-inflammatory IL-1ra in circulating plasma from a subset of the subjects in the original RCT. In both groups receiving 2'-FL-supplemented formula, compared to the control formula group, the levels of cytokines tested were lower and similar to those observed in breastfed infants. In supernatants of human peripheral blood mononuclear cell cultures following ex vivo stimulation by RSV44 viruses, the 2'-FL-supplemented formula group did not differ significantly from the breastfed group. There was also no difference between the two groups in any of the cytokines tested. However, the levels of some cytokines were higher in the control formula group. The clinical significance of the observed differences remains to be clarified.

2'-FL and LNnTSupplementation

A 2017 RCT by Puccio et al. [27] assessed the impact of infant formula supple-mented with 2'-FL and LNnT on growth, tolerance, and morbidity. Healthy infants born at term, recruited at the age of <14 days, were eligible for the study. Children meeting the inclusion criteria were randomly assigned to one of 2 groups in which they received intact-protein, cow's milk-based infant formula containing 2'-FL (1 g/L) and LNnT (0.5 g/L) (n = 88) or the same infant formula without supplementation (n = 87). The intervention lasted 6 months; then, all infants received standard follow-up formula without HMOs from 6 to 12 months. Baseline clinical and demographic characteristics did not differ significantly between the study groups. Both groups showed similar weight gain in the 4th month of life (primary endpoint) as well as similar other anthropometric parameters (body weight and length, head circumference, and BMI - all expressed as z-scores) measured throughout the observation period. Gastrointestinal tolerance was similar in both groups except for softer stool in the 2'-FL- and LNnT-supplemented formula group at 2 months. In addition, in the group fed with 2'-FL- and LNnT-supplemented formula, compared to the group fed with standard formula, there was a lower risk of parent-reported adverse events identified a priori and less medication use (secondary endpoints), including:
•    Lower risk of bronchitis throughout all examined time intervals (assessed at 0-4, 0-6, and 0-12 months of life)
•    Lower risk of lower respiratory tract infections (assessed at 0-12 months of life)
•    Lower use of antipyretics (assessed at 0-4 months of life)
•    Less use of antibiotics (assessed at 0-6 and 0-12 months of life)

The results regarding the lower risk of certain diseases and the lower use of medications require confirmation because they were parent-reported secondary endpoints (although verified by the study team). However, the impact of 2'-FL and LNnT supplementation on the use of antibiotics, which was reduced by 31% (relative risk 0.69, 95% CI 0.51-0.93), is important. Several studies have docu-mented the links between early-life antibiotic therapy and the risk of disease (including obesity and allergies) in later life [28-30]. Antibiotic administration by itself is neither necessary nor sufficient as a cause of these diseases. However, its potential role cannot be ignored, and every intervention which reduces the use of antibiotics is worth considering.

Later, it was reported that reduced overall medication use (antipyretics and antibiotics) may be linked to the gut microbiota types. In particular, formula-fed infants who demonstrated fecal community type with Bifidobacteriaceae at high abundance at 3 months were less likely to require antibiotics during the first year than those with fecal community type with less abundant Bifidobacteriaceae [31].

Protein Hydrolysates Supplemented with 2'-FL and LNnT

At the time of the writing of this chapter, the results of 2 trials published as full papers were available, in which 2'-FL and LNnT were added to partially or ex-tensively hydrolyzed formulas.

Whey-Based Partially Hydrolyzed Infant Formula with Bifidobacterium lactis and 2'-FL
In a 2019 RCT by Storm et al. [32], in which infants less than 2 weeks of age were eligible, the effect of using a whey-based (100%), partially hydrolyzed infant for-mula supplemented with Bifidobacterium animalis ssp lactis strain Bb12 (control group; n = 33) was assessed for 6 weeks compared to the same hydrolysate containing additionally 2'-FL (experimental group; n = 30). A similar tolerance was found for both preparations assessed using the Infant Gastrointestinal Symptom Questionnaire (IGSQ). This questionnaire included bowel movements, vomiting, spitting, crying, and fussiness.

Whey-Based Extensively Hydrolyzed Infant Formula Containing 2'-FL and LNnT
A 2019 clinical trial by Nowak-Wẹgrzyn et al. [33] involving 67 children aged 2 months to 4 years with cow's milk protein allergy confirmed the hypoallergenicity of a whey-based (100%) extensively hydrolyzed infant (test) formula containing 2'-FL and LNnT. All children were assessed by double-blind, placebo-controlled food challenges to the test and control formulas, in random order. The product is considered to meet hypoallergenic conditions if it is well tolerated by at least 90% of patients (with a 95% confidence interval). The outcome of the DBPCFC in response to the test and control formulas confirmed the hypoallergenicity of the test formula.

Infants with cow's milk protein allergy may be at higher risk of infections. The findings of the CINNAMON study (ClinicalTrials.gov Identifier: NCT03085134; so far only presented in abstract form) demonstrated that children with cow's milk allergy fed a whey-based (100%) extensively hydrolyzed infant formula containing 2'-FL and LNnT with a reduced protein content (2.2 g/100 kcal) compared with those fed an un-supplemented formula with a higher protein content (2.5 g/100 kcal) had similar growth and tolerated well both formulas. Children fed 2'-FL- and LNnT-supplemented formula appeared to have a reduced risk of respiratory infections and reduced use of antibiotics and antipyretics; however, the differences between groups were not significant.

Further Trials Are Needed and Are Underway

At the time of the writing of this chapter, there were at least 13 interventional studies registered in ClinicalTrials.gov to evaluate various structures identical to HMOs in the pediatric population. These studies were at various stages (i.e., re-cruiting, completed). Among them, there is an RCT evaluating a mixture of five of the most commonly occurring HMOs in breast milk, i.e. 2'-FL, 2',3-di-fu- cosyllactose, lacto-N-tetraose, 3'-sialyllactose, and 6'-sialyllactose. As no single ingredient is likely to act as a magic bullet, combining those with proven effectsin one product brings infant formula closer to human breast milk and is justified. However, safety and clinical effects are still to be documented.

To ensure the quality of infant nutritional studies, standards for conducting such studies have been in development [34-36]. The most recent example is the document published in JAMA Pediatrics [35]. One of the questions asked was how to ensure the validity of clinical trials of breast milk substitutes while protecting trial participants, particularly with regard to breastfeeding. In line with the document, if the trial aims to demonstrate adequate infant growth and tolerance of a new breast milk substitute, participating infants should be fully breast milk-substitute-fed, and the decision not to use breast milk should be firmly established prior to enrollment in the trial. Both intention-to-treat and prespecified per protocol analyses are recommended. If the trial aims to generate data to support a nutrition or health claim, some infants will be receiving breast milk at enrollment, and it is important to demonstrate adequate support for breastfeeding. Only intention-to-treat analysis is recommended. While not addressed in the above document, one important issue when designing the study is to define what is an important difference between groups, i.e. minimal clinically important difference (MCID). This is defined as “the smallest amount an outcome must change to be meaningful to patients” [37]. Trivial differences between groups can become statistically significant but should be interpreted with caution in the context of MCID.

Conclusions

Research on HMOs is one of the hot topics in infant nutrition. Progress in biotechnology nowadays allows the production of at least some structures identical to HMOs to supplement infant formulas. Overall, the findings from some recently published trials, along with the EFSA and FDA opinions, provide reassurance that infant formulas supplemented with selected structures identical to HMOs, specifically 2'-FL with/out LNnT, are safe and well tolerated, and may have favorable effects on some health outcomes and medication usage. Further studies are needed. While awaiting new evidence, it seems reasonable to discuss with care providers current evidence regarding HMOs and structures identical to HMOs added to infant formulas and let them decide whether the expected benefits are in line with their expectations and worth the costs incurred. Evidence presented in this review should facilitate such discussion.

Conflict of Interest Statement

H.S. has participated as a consultant and/or speaker for companies manufacturing infant formulas, i.e. Danone, Else Nutrition, Hipp, Mead Johnson/RB Health, Nestlé, Nestlé Nutrition Institute.

References
1    Victora CG, Bahl R, Barros AJ, et al. Lancet Breastfeeding Series Group. Breastfeeding in the 21st century: epidemiology, mechanisms, and lifelong effect. Lancet. 2016;387:475-90.
2    Ballard O, Morrow AL. Human milk composition: nutrients and bioactive factors. Pediatr Clin North Am. 2013;60:49-74.
3    Triantis V, Bode L, van Neerven RJJ. Immunological effects of human milk oligosaccharides. Front Pediatr. 2018;6:190.
4    Walsh C, Lane JA, van Sinderen D, Hickey RM. Human milk oligosaccharides: shaping the infant gut microbiota and supporting health. J Funct Foods. 2020 Sep;72:104074.
5    Moore RE, Xu LL, Townsend SD. Prospecting human milk oligosaccharides as a defense against viral infections. ACS Infect Dis. 2021 Feb;12;7(2):254-63.
6    Morrow AL, Ruiz-Palacios GM, Altaye M, et al. Human milk oligosaccharides are associated with protection against diarrhea in breast-fed infants. J Pediatr. 2004 Sep;145(3):297-303.
7    Seppo AE, Autran CA, Bode L, Jarvinen KM. Hu-man milk oligosaccharides and development of cow's milk allergy in infants. J Allergy Clin Im-munol. 2017 Feb;139(2):708-11.e5.
8    Miliku K, Robertson B, Sharma AK, et al. CHILD Study Investigators, Bode L, Azad MB. Human milk oligosaccharide profiles and food sensitiza-tion among infants in the CHILD Study. Allergy. 2018 Oct;73(10):2070-73.
9    Lodge CJ, Lowe AJ, Milanzi E, et al. Human milk oligosaccharide profiles and allergic disease up to 18 years. J Allergy Clin Immunol. 2021 Mar;147(3):1041-48.
10    Sprenger N, Odenwald H, Kukkonen AK, et al. FUT2-dependent breast milk oligosaccharides and allergy at 2 and 5 years of age in infants with high hereditary allergy risk. Eur J Nutr. 2017 Apr;56(3):1293-301.
11    Autran CA, Kellman BP, Kim JH, et al. Human milk oligosaccharide composition predicts risk of necrotising enterocolitis in preterm infants. Gut. 2018 Jun;67(6):1064-70.
12    Masi AC, Embleton ND, Lamb CA, et al. Human milk oligosaccharide DSLNT and gut microbiome in preterm infants predicts necrotising enteroco-litis. Gut. 2020 Dec 16:gutjnl-2020-322771.
13    Torres Roldan VD, Urtecho SM, Gupta J, et al. Human milk oligosaccharides and their association with late-onset neonatal sepsis in Peruvian very-low-birth-weight infants. Am J Clin Nutr. 2020 Jul 1;112(1):106-12.
14    Larsson MW, Lind MV, Laursen RP, et al. Human milk oligosaccharide composition is associated with excessive weight gain during exclusive breastfeeding-an explorative study. Front Pedi- atr. 2019 Jul 18;7:297.
15    Lagstrõm H, Rautava S, Ollila H, et al. Associations between human milk oligosaccharides and growth in infancy and early childhood. Am J Clin Nutr. 2020 Apr 1;111(4):769-78.
16    Salminen S, Stahl B, Vinderola G, Szajewska H. Infant formula supplemented with biotics: current knowledge and future perspectives. Nutrients. 2020 Jun 30;12(7):1952.
17    Walsh C, Lane JA, van Sinderen D, Hickey RM. From lab bench to formulated ingredient: charac-terization, production, and commercialization of human milk oligosaccharides. J Funct Foods. 2020;72:104052.
18    EFSA NDA Panel (EFSA Panel on Dietetic Prod-ucts, Nutrition and Allergies), 2015. Scientific opinion on the safety of 2'-O-fucosyllactose as a novel food ingredient pursuant to Regulation (EC) No 258/97. EFSA J 2015;13(7):4184.
19    http://www.accessdata.fda.gov/scripts/ fdcc/?set=GRASNotices
20    Bych K, Miks MH, Johanson T, et al. Production of HMOs using microbial hosts - from cell engi-neering to large scale production. Curr Opin Bio- technol. 2019Apr;56:130-7.
21    Akkerman R, Faas MM, de Vos P. Non-digestible carbohydrates in infant formula as substitution for human milk oligosaccharide functions: effects on microbiota and gut maturation. Crit Rev Food Sci Nutr. 2019;59(9):1486-97.
22    Phipps KR, Baldwin N, Lynch B, et al. Safety eval-uation of a mixture of the human-identical milk oligosaccharides 2'-fucosyllactose and difucosyl- lactose. Food Chem Toxicol. 2018 Oct;120:552- 65.
23    Morozov V, Hansman G, Hanisch FG, et al. Human milk oligosaccharides as promising antivirals. Mol Nutr Food Res. 2018 Mar;62(6):e1700679.