Optimization of metagenomic detection method for human breast milk microbiome

Abstract

This study aims to optimize the metagenomic detection methodology of the human breast milk microbiome and analyze its composition. Twenty-two milk samples were collected from the left and right sides of lactating women during re-examinations at the Haidian Maternal and Child Health Hospital, Beijing. Microbial cell wall disruption parameters were optimized, and a nucleic acid extraction method was developed to construct a microbial DNA/RNA library. Metagenomic next-generation sequencing (mNGS) sequencing was performed, and microbial composition was analyzed using the k-mer Lowest Common Ancestor (LCA) method with a self-generated database constructed via Kraken2 software. Data showed Q20 > 95% and Q30 > 90%, with an average total data volume of 5,567 ± 376.6 Mb and non-human sequence data of 445.1 ± 63.75 Mb, significantly enhancing sequencing efficiency. The microbiome included 21 phyla, 234 genera, and 487 species, with Firmicutes and Proteobacteria as dominant phyla. Predominant genera included Staphylococcus and Streptococcus, and major species were Staphylococcus aureus, Streptococcus bradystis, and Staphylococcus epidermidis. Species levels exhibited significant variations among different individuals. Microbial profiles of left- and right-sided milk samples were consistent at the phylum, genus, and species levels. In addition to common bacteria, diverse viral, eukaryotic, and archaeal sequences were detected. This study refined metagenomic detection methods for human breast milk microbiota. Specific flora colonization occurred in healthy breast milk, with the left and right sides exhibiting both correlations and distinct flora environments.