Characterization of human breast tissue microbiota from core needle biopsies through the analysis of multi hypervariable 16S-rRNA gene regions

Abstract

Breast microbiota compositions are not well understood, and a few recent reports have begun to explore the correlation between breast tissue dysbiosis and cancer. Given that various methods for breast microbiota detection were used, the aim of the present paper was to clarify which hypervariable region of the 16S-rRNA gene (V2, V3, V4, V6 + 7, V8, and V9) is the most informative for breast tissue microbiota. Core needle biopsies (CNBs) were compared with surgical excision biopsies (SEBs) to find a less invasive form of recovery useful for the analysis of a larger statistical population and potentially for diagnostic use of breast tissue microbiota. Finally, this study was the first to analyse the breast microbiota of tumours and paired normal tissues of a Mediterranean population. Our findings showed that the V3 region is the most informative for breast tissue microbiota, accounting for 45% of all reads. No significant differences were found between CNB and SEB specimens in terms of total reads and numbers of Operational Taxonomic Units (OTUs). Moreover, we find that more similarities than differences exist between tumours and adjacent normal tissues. Finally, the presence of the Ralstonia genus is associated with breast tissue.

Figure 1

Comparison of the efficiency of analysed hypervariable regions of the 16S-rRNA gene (n = 32). (a) Distribution of obtained total read counts for each analysed hypervariable region. (b) Evaluation of read distributions of the analysed 16S-rRNA gene hypervariable regions and of each phylum found to compare the informative power of the regions. In both panels values are presented as relative abundances as percentages of total read counts obtained by combining data drawn for all of the analysed specimens (both SEBs and CNBs, healthy and cancerous).

Figure 2

Alpha-diversity analysis after rarefaction (n = 30). A comparison of the two alpha-diversity measures for the two tissue and sampling types: (a) number of observed OTUs, (b) Shannon index.

Figure 3

Dissimilarity analysis (beta-diversity) of the healthy and cancerous CNB and SEB samples. (n = 30). (a) The weighted UniFrac distance was used to analyse distances between cancerous and healthy samples of different subjects (between subjects, BS) in comparison to healthy and cancerous samples of the same subject (within subject, WS). (b) Principal coordinate analysis (PCoA) based on Weighted UniFrac distances.

Figure 4

Relative abundances measured at the genus and family levels. Barplots of the taxonomic profiles with each bar representing a subject and with each coloured box showing a bacterial taxon. (a) Barplots of each specimen measured at the genus level; (b) barplots of each specimen measured at the family level.