The My Baby Biome study is a seven-year longitudinal initiative focusing on the health and gut microbiome composition of infants in the United States. This innovative research method employs a decentralized clinical trial approach to reflect the diverse demographics of the U.S. population.
Birth modes and feeding methods among study participants were found to be comparable to data from the Centers for Disease Control and Prevention (CDC). This included both vaginal and C-section births, as well as exclusively breastfed, formula-fed, and combination-fed infants.
Fecal samples were collected from participants across 48 of the 50 states. Initial analyses were conducted on the microbiome during a critical period for immune development, specifically when infants were just one to three months old.
Metagenomic and metabolomic analyses revealed that 559 distinct species of bacteria were present, with an average of 12.1 species identified per sample. Notably, the infant gut microbiome demonstrated significant simplification compared to adult microbiomes.
Bifidobacterium species emerged as predominant in the dataset, underscoring their importance. However, a concerning bimodal distribution was identified: 24% of infants, including 19% of those born vaginally and 35% by C-section, lacked observable Bifidobacterium altogether. When present, these bacteria were found in substantial quantities.
Among the most abundant species were Bifidobacterium breve, Bifidobacterium bifidum, Bifidobacterium longum subsp. longum, and Bifidobacterium longum subsp. infantis. B. longum subsp. infantis was particularly rare, appearing in only 8% of the samples.
Feeding methods and birth modes significantly impacted the composition of the infant microbiome. Interestingly, the study showed that there were no significant trends in microbiome composition based on other factors, such as age, gender, race, or geographic location.
A deeper analysis revealed differences in species prevalence based on birth mode, with all species more common in vaginal born infants belonging to the phylum Bacteroidota. In contrast, infants born via C-section tended to have higher levels of Firmicutes.
Though Bifidobacterium were not directly associated with birth mode, their presence could occur through environmental exposure. The study emphasized that known human milk oligosaccharide (HMO) consumers like B. infantis had higher abundance in breastfed infants, compared to those fed formula or a mixed diet.
Concerningly, potentially pathogenic bacteria such as Klebsiella pneumoniae and Clostridium perfringens were also more abundant in the breastfed group, indicating their ability to thrive in the HMO-consuming environment.
A remarkable finding was the impact of the synergistic relationship between birth mode and feeding method. Breastfeeding was related to increased Bifidobacterium levels in vaginal birth infants but showed a reverse trend in C-section infants, where breastfeeding associated with a decrease in these beneficial microbes.
To explore these dynamics further, researchers analyzed bacteria that increased in the absence of Bifidobacterium. The most significantly noted bacteria among these was Clostridium perfringens, recognized as a potential pathogen with known capabilities of HMO utilization.
Using Dirichlet Multinomial Mixture (DMM) model clustering, the researchers categorized the samples into three groups representing different microbial compositions: C1 (24%), C2 (37%), and C3 (39%). These clusters were found to correlate strongly with birth modes and feeding methods, revealing a clear connection to health outcomes.
The research further highlighted that the lack of Bifidobacterium largely drove microbial shifts, with a specific emphasis on how these shifts related to functional analysis, including HMO utilization capabilities, antimicrobial resistance genes, and virulence factors across the dataset. These functional perturbations appear critical as HMOs are vital for immune development and microbiome maturation.
The presence of Bifidobacterium was confirmed as significant for HMO utilization, particularly B. infantis. The analysis discovered that 97% of samples within Cluster C1 had Bifidobacterium as the primary HMO contributor, while this declined significantly in Cluster C2 and was replaced by Clostridia in Cluster C3.
Investigating broader carbohydrate utilization patterns, the researchers looked at fucosidases and sialidases to determine how different clusters managed HMO utilization. The findings indicated that while C1 exhibited robust HMO utilization, C2 and C3 were functionally limited, indicating potential dysbiosis associated with the lack of Bifidobacterium.
Further investigations into antimicrobial resistance genes and virulence factors linked these elements to the abundance of Bifidobacterium. Surprisingly, higher relative abundances of AMR genes were observed in vaginal births, while C-section infants had elevated levels of virulence factor genes.
A significant relationship was noted showing that as the abundance of Bifidobacterium decreased, the presence of AMR and virulence factor genes increased. Additionally, the highest relative abundances of AMR and virulence factor genes were identified in cluster C3, demonstrating the ramifications of lacking beneficial gut microbes in infancy.
To examine the metabolic impact of Bifidobacterium on the infant gut microbiome, researchers quantified 79 metabolites in infant fecal samples. Feeding mode was identified as a confounder, notably influencing 29 metabolites.
Significant metabolic differences were identified among the clusters, particularly in Cluster C3, which exhibited altered bile acid metabolism and reduced thiamine production. Furthermore, variations in short-chain fatty acid profiles were noted, with higher levels of butyrate being produced in Cluster C3.
To assess host-microbiome interactions, a microbe-metabolite co-occurrence network was constructed, revealing complex interconnections. This analysis underscored the complementary roles of Bifidobacterium species, particularly in suppressing pathogens and enhancing beneficial metabolites.
As part of the ongoing assessment, health outcome data was collected at the two-year mark. Among 412 initial participants, about 53.8% of parents reported antibiotic use between birth and two years of age.
30% of parents reported adverse health outcomes, including allergies, asthma, and eczema reported diagnoses. In examining associations based on microbial clusters, infants in clusters C2 and C3 exhibited significantly higher likelihoods of developing adverse outcomes, raising concerns about these particular community states.
Notably, higher relative risks for adverse health outcomes were also linked to antibiotic usage. Analysis revealed a notable protective effect stemming from the presence of Bifidobacterium, indicating their role in mitigating adverse health risks in this population.
A complex analysis using ortholog clustering approached further revealed associations between function and adverse health outcomes. Key gene clusters linked to increased risk were identified, emphasizing the need for further validation.
The My Baby Biome study aims to continue monitoring health trends for seven years, with alarming statistics already surfacing by the two-year mark regarding the vulnerable population’s health. The observation that a significant chunk of infants lack critical Bifidobacterium emphasizes a pressing issue that merits further research.
The absence of these beneficial microbes corresponds with the emergence of potentially opportunistic and pathogenic bacteria which heightens the risk of health issues. The analyses presented underscore the functional changes the infant gut microbiome undergoes in this context, aligning with known detrimental health outcomes.
As nutritional practices and antibiotic use have radically transformed in modern society, understanding their role in early microbiome development offers potential pathways for early interventions that may positively influence lifelong health outcomes.
image source from：nature
