Expression and cold storage of human milk is a common practice. However, the effect of typical cold storage on the human milk microbiome is not well understood. Here, the effect of cold storage on the human milk microbiome was investigated. In order to assess the both the total bacterial DNA profile and the viable microbiome of human milk, propidium monoazide (PMA) coupled with full-length 16S rRNA gene sequencing was used, allowing differentiation between DNA from live and dead bacteria.
Human milk samples (n=10) were collected and immediately transported to the laboratory for processing (<30 minutes). Each sample was split into 10 x 1 ml aliquots and either processed immediately (fresh), or stored at 4⁰C for 4 days, -20⁰C for 2 months, -20⁰C for 6 months, or -80⁰C for 6 months. At each time point two aliquots from each sample were analysed: one with a PMA pre-treatment (to analyse DNA from viable bacteria only), and one without.
PMA-treated samples differed significantly in their bacterial composition from untreated samples (AMOVA p=0.001). While untreated samples had an average richness (number of OTUs) of 55, PMA-treated samples had an average richness of 18 (p<0.0001). Additionally, numerous taxa were significantly less abundant in PMA-treated vs untreated samples. Overall, cold storage did not significantly affect either the total bacterial DNA profile or the viable microbiome of human milk.
These findings raise questions about the validity of conclusions drawn from previous milk microbiome studies which have used 16S rRNA gene sequencing without PMA, as the bacterial DNA content of milk does not reflect the viable bacterial content. However, the presence of dead bacteria in milk may also serve a biological function for the infant, such as immune training. Our finding that cold storage does not impact the human milk microbiome is reassuring for researchers who may store samples prior to analysis, and for mothers who may store expressed breast milk.