BACKGROUND: Observational epidemiological studies suggested an association between the gut microbiota and breast cancer, but it remains unclear whether the gut microbiota causally influences the risk of breast cancer. We employed two-sample Mendelian randomization (MR) analysis to investigate this association. METHODS: We used summary statistics of the gut microbiome from a genome-wide association study (GWAS) of 18,340 individuals in the MiBioGen study. GWAS summary statistics for overall breast cancer risk and hormone receptor subtype-specific analyses were obtained from the UK Biobank and FinnGen databases, totaling 400,000 individuals. The inverse variance-weighted (IVW) MR method was used to examine the causal relationship between the gut microbiome and breast cancer and its subtypes. Sensitivity analyses were conducted using maximum likelihood, MR-Egger, and MR pleiotropic residual sums and outliers methods. RESULTS: The IVW estimates indicated that an increased abundance of Genus_Sellimonas is causally associated with an increased risk of ER+ breast cancer [odds ratio (OR) = 1.09,  = 1.72E-04, false discovery rate (FDR) = 0.02], whereas an increased abundance of Genus_Adlercreutzia was protective against ER+ breast cancer (OR = 0.88,  = 6.62E-04, FDR = 0.04). For Her2+ breast cancer, an increased abundance of Genus_Ruminococcus2 was associated with a decreased risk (OR = 0.77,  = 4.91E-04, FDR = 0.04), whereas an increased abundance of Genus_Erysipelatoclostridium was associated with an increased risk (OR = 1.25,  = 6.58E-04, FDR = 0.04). No evidence of heterogeneity or horizontal pleiotropy was found. CONCLUSION: Our study revealed a gut microbiota-mammary axis, providing important data supporting the potential use of the gut microbiome as a candidate target for breast cancer prevention, diagnosis, and treatment.