1. Introduction
Human milk oligosaccharides (HMOs) are a structurally diverse group of carbohydrates exclusive to human milk. In the composition of human milk’s non-liquid components, HMOs rank third in abundance, following closely behind fat and lactose. These bioactive substances play a critical role in shaping the robust growth patterns and facilitating the various stages of infant development throughout the first few years [
1,
2,
3,
4]. Currently, researchers have discovered over 200 molecular structures of HMOs in breast milk. HMOs consist of five monomers: glucose (Glc), galactose (Gal), N-acetylglucosamine (GlcNAc), fucose (Fuc), and N-acetylneuraminic acid (Neu5Ac). Moreover, all HMOs have lactose (Gal-β-1,4-Glc) at their reducing end [
4]. In spite of the considerable heterogeneity in their molecular structures, a tripartite classification system exists for HMOs depending on the essence of their central structure, encompassing fucosylated neutral types, non-fucosylated neutrals, and sialylated species [
5]. HMOs are primarily resistant to direct digestion and absorption within the human body, making them unavailable for immediate utilization by infants. Nevertheless, HMOs have the ability to reach the colon directly and serve as a source of nourishment for specific microorganisms within the infant’s gut [
6,
7,
8]. A significant observation to make is the relatively diminished extent to which the resident gut microbes effectively utilize and metabolize HMOs [
9].
In the initial stages of life,
Bifidobacterium assume a preeminent position within the infant gut microbiome, exhibiting a strong correlation with substantial shifts in the composition and structure of the intestinal microbial community during this critical period [
10,
11]. When compared to the adult intestine, the intestinal microbiota of infants exhibits lower diversity and a more unstable microbial structure. Currently, studies suggest that early-life enterotypes can be categorized into four groups, with the key microorganisms being Firmicutes,
Bifidobacterium,
Bacteroides, and
Prevotella. As infants grow, the first two microorganisms replace the latter, and the development process of their intestinal microbiota becomes deterministic and predictable [
12]. The establishment of early-life intestinal microbiota is directly linked to health risks at each developmental stage [
13,
14,
15].
In light of the extensive array of benefits it imparts, the World Health Organization (WHO) advocates for a period of exclusive breastfeeding extending from birth through the sixth month of infancy for both the baby’s and mother’s well-being [
16,
17]. Among these benefits, HMOs play a regulatory role in the infants’ intestinal tract, promoting the growth of
Bifidobacterium. However, the digestive capabilities of HMOs exhibit strain specificity [
18,
19].
Bifidobacterium longum subsp. infantis (
B. infantis),
Bifidobacterium bifidum (
B. bifidum),
Bifidobacterium longum subsp. longum (
B. longum), and others can utilize HMOs in the gut. HMOs undergo specific microbial metabolism in the gut, producing short-chain fatty acids (SCFAs). The intestinal epithelium utilizes SCFAs as an energy source, which in turn supports and optimizes the performance of the intestinal barrier system [
20,
21].
In vitro simulated fermentation systems of specific prebiotics can be employed to assess the interaction of microbiota in the intestine and the fermentation characteristics of prebiotics [
22,
23]. Currently, fermentation methods applied to fecal bacteria cultures are classified into two categories: batch cultures and continuous cultures. Batch cultures are widely used because of their convenient operation and the quick assessment of the intestinal microbiota’s ability to utilize specific carbon sources [
24]. In this study, the yeast extract–casein hydrolysate–fatty acids medium (YCFA) was used, capable of culturing the majority of the microbiota in the intestine. In a study, fecal samples were harvested from six healthy volunteers and utilized for both YCFA-based microbial cultivation and extensive metagenomic sequencing. The outcome disclosed that, on a mean basis, about 93% of the raw sequencing outputs remained consistent between the native fecal microorganisms and their corresponding cultured counterparts among all six contributors [
25].
The primary goal of this study was to quantify the abundance of 10 Bifidobacteria species in the infant gut microbiota and evaluate the in vitro fermentation outcomes of the infant gut microbiota where Bifidobacterium served as the dominant genus in the presence of specific HMOs. At the outset, we recruited a group comprising 41 infants within their first 6 months of life, all born at full term. Utilizing data from a foundational 16S rRNA sequencing examination, we then proceeded to select samples with a relative Bifidobacterium abundance exceeding 60% for further investigation through an in vitro manipulation. The selected HMOs included neutral fucosylated HMO (2′-fucosyllactose, 2FL), sialylated HMOs (3′-sialyllactose, 3SL; 6′-sialyllactose, 6SL), and nonfucosylated neutral HMOs (lacto-N-tetraose, LNT; lacto-N-neotetraose, LNnT). Simultaneously, the selected prebiotics currently added to infant formulae, namely Galactooligosaccharides (GOSs) and Fructooligosaccharides (FOSs), were used as positive controls. The contents of 10 special Bifidobacteria species and SCFAs were determined after 24 h of fermentation, providing the possibility of adding HMOs to formula milk powder.
4. Discussion
The current study aimed at investigating the potential of five different commercial HMOs, including 2FL, 3SL, 6SL, LNT, and LNnT, each characterized by specific structural configurations, to act as the sole carbon sources during the in vitro fermentation by the complex microorganisms within the infant gut ecosystem. In order to assess how distinct HMOs influence the structure and diversity within the infant’s gut microbial community, methodologies such as 16S rRNA sequencing, alongside qPCR, were utilized in our study. The focus was on examining alterations in the dominant genera and the 10 species of Bifidobacteria. In addition, changes in the content of SCFAs after fermentation of different HMOs were analyzed and correlated with gut microbiota.
Findings from our 16S rRNA sequencing data indicated that the capacity for assimilating HMOs among the gut bacteria did not rely on the prevalence or domination of specific species. Instead, this ability seemed more closely related to the initial makeup of the resident intestinal microbiome. The species within dominant genera remained unchanged after 24 h of fermentation by HMOs with different structures, yet their quantity and distribution underwent significant alterations. The relative abundance of beneficial bacteria, such as
Bifidobacterium and
Lactobacillus, increased, while that of potentially harmful bacteria like
Escherichia and
Enterococcus decreased. These findings align with previous studies [
34,
35,
36]. In contrast to the well-documented HMO utilization strategies of
Bifidobacteria, the molecular mechanisms by which
Lactobacillus species process HMOs remain relatively uncharted. Prior research has indeed shed some light on this subject through genomic analyses of
Lactobacillus strains, revealing that they possess a rather constrained capacity for fermenting HMOs [
37,
38]. Recent research findings have illuminated that the lactose manipulator enzyme in
Lactobacillus casei plays a pivotal role in the transportation and metabolic processing of core-2 N-acetyllactosamine, a key component found within HMOs [
39]. This suggests that
Lactobacillus casei degrades HMO through different metabolic pathways. Moreover, LEfSe analysis revealed a more robust probiotic function associated with HMOs. Species that significantly differed in the HMO group based on LDA scores included
Bacteroides.
Bacteroides are known to employ various strategies for degrading HMOs [
40]. Notably,
Akkermansiaceae emerged among the significantly different species associated with HMOs. This family, belonging to Verrucomicrobia and represented by a single member,
Akkermansia muciniphila [
AKK], is recognized as a potential probiotic. Studies propose that
AKK can thrive on breast milk by utilizing HMOs [
41].
This study revealed that HMOs promote the growth of beneficial bacteria, particularly
Bifidobacterium. Hence, HMOs were initially dubbed the “bifidus factor” in breast milk. Among the 10 specific
Bifidobacterium species studied, strain specificity was evident in their utilization of HMOs.
B. infantis and
B. bifidum displayed an ability to utilize multiple types of HMOs effectively, with 6SL significantly boosting the growth of both (
p < 0.05). In contrast, the remaining seven
Bifidobacterium strains—
B. longum,
B. dentium,
B. thermophilum,
B. animalis,
B. adolescentis,
B. angulatum, and
B. pseudocatenulatum—were unable to utilize a variety of HMO species. Notably,
B. longum,
B. pseudocatenulatum,
B. adolescentis, and
B. angulatum exhibit the capacity to utilize sialylated HMOs or their degradation products for enhanced metabolic value. Among these, only
B. breve exhibited a minimal capacity for HMO degradation, differing from previous studies [
42].
The complete breakdown of these HMOs, which come in diverse molecular structures, necessitates the intricate interplay of a series of glycoside hydrolases or membrane transport proteins that are present within the infant gut environment. A majority of
Bifidobacterium strains have evolved to genetically encode specific enzymes for HMO degradation and transporter proteins, enabling them to selectively utilize HMOs with distinct structures. In the infant’s gut,
Bifidobacterium employs two distinct strategies for HMO utilization: intracellular digestion and extracellular digestive mechanisms.
B. infantis can transport HMOs into the cytoplasm through various oligosaccharide transport proteins. Subsequently, intracellular glycosyl hydrolases (GHs) play a role in the degradation of most HMOs [
43]. This microorganism also demonstrates a unique extracellular tactic in which HMOs are enzymatically cleaved outside the cell by specific glycosidases into simpler sugars, such as monosaccharides or disaccharides. Subsequently, these hydrolyzed compounds are absorbed into the bacterial cell’s cytoplasm to be metabolized further [
40]. Conversely,
B. bifidum and some
B. longum utilize extracellular GHs for the initial degradation of HMOs. Subsequently, they transport the intermediates, along with specific transport proteins, into the cell for further catabolic utilization [
33].
Indeed, there may be a synergistic effect between
Bifidobacteria. Studies indicate that
B. bifidum SC555 may release fucose and sialic acid into the environment during its growth on short-chain HMOs, potentially cross-feeding other
Bifidobacteria [
44]. Reflecting on our study results, we observed growth in
B. adolescentis,
B. dentium,
B. angulatum,
B. pseudocatenulatum,
B. thermophilum, and
B. animalis within the HMO group. This growth could be attributed to such cross-feeding or other metabolic adaptations, as previous studies have shown that these species either failed to grow or produced only minimal metabolites in response to HMO stimulation [
45,
46,
47]. However, it has been suggested that
B. dentium possesses the ability to remove terminal fucose from HMOs [
48]. Researchers found a smaller number of glycosyl hydrolase genes in the genomes of
Bifidobacterium angulatum JCM 7096(T) and
Bifidobacterium pseudocatenulatum JCLA3 [
49,
50]. The
B. thermophilum genome contains a β-galactosidase gene, a finding that hints at its potential to degrade HMOs since β-galactosidase can participate in their metabolism; however, the exact mechanism remains uncharted [
50].
The production of SCFA stands as a crucial physiological process orchestrated by gut microbiota. This process contributes to sustaining a low-pH environment in the intestine and plays a vital role in connecting microbial communities with host immunity [
40,
51]. Within HMOs, sialylated HMOs exhibited the highest acid production capability, while neutral fucosylated HMOs and FOSs demonstrated comparable acid production levels. Interestingly, GOSs in the positive control generated the least acid, deviating from findings in previous studies [
1,
52]. Furthermore, our study, employing
B. infantis-dominant and
B. breve-dominant strains as inocula, demonstrated varying levels of AA production by
Bifidobacteria during HMO degradation. Specifically, the
B. infantis-dominant strain exhibited significantly higher levels of AA production than the
B.breve-dominant strain (
p = 0.000) (
Figure A2). There is compelling evidence from studies indicating that among children at elevated risk for asthma, a common observation is lower intestinal abundance of AA, which may allude to an underlying association between AA and the progression of allergic pathologies in early life [
53]. Martin and colleagues observed a direct correlation between an increase in the abundance of
Bifidobacterium and a concomitant rise in AA content. This finding suggests that the proliferation of
Bifidobacterium is positively associated with enhanced AA production within the gut environment [
54]. Interestingly, there was minimal production of PA and BA at the conclusion of fermentation. Conversely, in a separate study using
Bacteroides-dominated inocula, PA and BA were generated in the later phases of fermentation [
52]. This discrepancy may stem from distinct HMO utilization strategies employed by
Bifidobacterium and
Bacteroides. This observation highlights that distinct initial microbial compositions result in varying metabolites. While the preliminary data suggest a potential link, establishing a definitive correlation between the initial gut microbiota composition and the metabolic fingerprints that emerge following HMO fermentation would require a considerably wider and more inclusive research cohort, thus allowing for more precise insights into this intricate dynamic.
During the ultimate phase of our study, we delved into the interplay between SCFAs and the gut microbiota prior to and following fermentation. The transition in the correlation patterns that occurred is plausibly connected to the reconfiguration of the gut microbiota’s composition and its altered metabolic behavior as a result of the fermentation process [
55]. As expected and consistent with prior research [
56], the correlation between
Bifidobacteria and SCFAs strengthened following 24 h of fermentation. Nevertheless, at the species-level of
Bifidobacteria, the robust negative correlation between microbiota and metabolites was diminished through HMO fermentation. Notably, the correlation of
B. infantis and
B. adolescentis with AA was also modified. In conclusion, the inclusion of prebiotics in the diet leads to substantial alterations in the composition and metabolic functionality of the gut microbiota, highlighting their pivotal role in modulating the gut environment. Notably, a separate study demonstrated a significant association between
B. pseudocatenulatum and
B. longum with AA in the gut of 6-year-old children, whereas
B. adolescentis exhibited a significant association with AA in the gut of adults [
57]. Interestingly, these findings differ from the results of the present study, suggesting distinct correlations between microbiota and metabolites in the gut among various age groups. Therefore, it is recommended to tailor nutritional supplements according to the specific interplay between distinct microbiota and metabolites, considering the variations across different age groups.
Several limitations characterize this study. Firstly, the modest sample size constrained our ability to draw conclusions from a diverse array of infant subjects. To mitigate potential biases stemming from individual variability, it is essential to expand our research by recruiting a larger population for future studies. Secondly, we recognized that the analysis was confined to a select few microbial metabolites, thus impeding a thorough comprehension of the multifaceted connections between the diverse microbial populations and their associated metabolic outputs. A more holistic approach would involve subjecting both pre- and post-fermentation fluids to extensive metabolomic analysis to uncover hidden patterns and connections.
5. Conclusions
In summary, the fermentation of HMOs by the infant gut microbiota did not alter the dominant genera within the microbial population. However, it induced significant changes in the composition and distribution of these microorganisms. Broadly speaking, there was a notable increase in the relative abundance of beneficial bacteria, particularly Bifidobacterium species. Among the 10 specific Bifidobacterium species studied, strain specificity was evident in their utilization of HMOs. B. infantis and B. bifidum displayed an ability to utilize multiple types of HMOs effectively, with 6SL significantly boosting the growth of both (p < 0.05). In contrast, the remaining Bifidobacterium strains were unable to utilize a variety of HMO species. Post-fermentation, acetic acid (AA) content increased significantly (p < 0.001), reaching levels approximately between 500 and 1500 μg/g. Conversely, the contents of propionic acid (PA, p = 0.314), butyric acid (BA, p = 0.872), and valeric acid (VA, p = 0.052) remained largely unchanged. Using fecal inocula dominated by either B. infantis or B. breve, our results demonstrated that different dominant microorganisms led to varying levels of AA production. Specifically, the B. infantis-dominant inoculum generated significantly higher amounts of AA compared to the B. breve-dominant one (p = 0.000). Ultimately, we underscored that the fermentation characteristics of the five HMOs were more closely related to the initial gut microbial composition rather than the degree of dominance of any particular species.
Our findings highlight the crucial importance of the initial microbial composition of fecal donors in shaping fermentation outcomes. This insight paves the way for the development of personalized formulas tailored to infants with specific dominant microbial profiles, allowing for the optimization of nutritional conditions for each infant group. Looking forward, it is envisioned that future advancements will facilitate the production of HMO-enriched formulas that are more aptly aligned with the gut health needs of the majority of infants, regardless of the specific donor origins.