Early development of infant gut microbiota in relation to breastfeeding and human milk oligosaccharides

**Background:** The infant gut microbiota plays a critical role in growth, immune development, and long-term health. Breastfeeding is a major determinant of gut microbiota composition, partly through human milk oligosaccharides (HMOs), which act as selective prebiotics for Bifidobacterium. However, clinical evidence linking specific HMO levels to infant gut microbiota composition remains limited. This study aimed to investigate associations between infant gut microbiome development, infant age, exclusive breastfeeding (EBF) duration, and HMO composition in breastmilk.

**Methods:** A total of 94 healthy mother-infant pairs were recruited from the Cambridge Baby Growth and Breastfeeding Study (CBGS-BF) in Cambridge, UK. All infants were singletons, vaginally born at term from healthy mothers with normal pre-pregnancy BMI. Infants were categorized into three feeding groups based on EBF duration: <3 months (n=13), 3–6 months (n=51), and >6 months (n=27). Infant stool samples (n=337) were collected at 2 weeks, 6 weeks, 3 months, and 6 months of age. The V3-V4 region of the 16S rRNA gene was sequenced using Illumina MiSeq. Mother's hindmilk samples were collected at birth, 2 weeks, 6 weeks, 3 months, and 6 months postpartum. Concentrations of five neutral HMOs (2'FL, 3'FL, LNFP1, LNnT, LNT) and two acidic HMOs (3'SL, 6'SL) were measured using HPAEC-PAD. Generalized linear mixed models (gLMM) and PERMANOVA were used for statistical analyses, with FDR correction for multiple comparisons.

**Key Results:** Bifidobacterium was the most abundant genus at all time points irrespective of breastfeeding duration, with an overall mean relative abundance of 70%. Bifidobacterium relative abundance increased significantly over time (p<0.001), from approximately 50% at 2 weeks to ~75% at 6 months. B. bifidum abundance was significantly higher in infants EBF >6 months compared to those EBF <3 months (p=0.0285). Alpha-diversity showed a biphasic pattern: Shannon index decreased by 0.13 units from 2 to 6 weeks (p=0.0068) and by 0.14 units from 6 weeks to 3 months (p=0.0022), then increased by 0.15 units from 3 to 6 months (p=0.00026). Richness decreased by 2.97 ASVs from 2 to 6 weeks (p=0.0022), was unchanged from 6 weeks to 3 months (p=0.50), then increased by 8.48 ASVs from 3 to 6 months (p=2.7×10⁻⁸). Richness was significantly lower in longer EBF groups (approximately 12 fewer ASVs in the 6 weeks–3 months period, p=0.01). Beta-diversity analysis showed that most variance was explained by inter-individual differences (R²=0.556, p<0.001) and time point (R²=0.06, p<0.001). Higher levels of 2'FL and LNFP1 in breastmilk were significantly associated with higher Bifidobacterium relative abundance across all time points (p=0.049 and p=0.017, respectively), while higher LNT was associated with lower Bifidobacterium abundance (p=0.029). A trend toward higher B. longum with higher LNFP1 was observed (p=0.078). No significant associations were found for LNnT, 3'SL, or 6'SL with Bifidobacterium.

**Clinical Implications:** This study provides direct evidence that EBF duration during the first months of life significantly impacts infant gut microbiota composition, with longer EBF associated with higher B. bifidum abundance and lower microbial richness. The significant associations between specific HMOs (2'FL and LNFP1) and Bifidobacterium abundance demonstrate a mechanistic pathway through which breastmilk components shape the infant microbiome. These findings support the importance of promoting exclusive breastfeeding and suggest that HMO profiles in breastmilk may be a target for interventions aimed at supporting healthy gut microbiota development in early life. Limitations include the inability to distinguish B. longum subspecies via 16S sequencing, analysis of only seven HMOs, and lack of data on actual HMO intake volume.