Validation of RNA isolated from milk fat globules to profile mammary epithelial cell expression during lactation and transcriptional response to a bacterial infection

Mastitis, an inflammation of the mammary gland, is the most costly infectious disease of dairy ruminants worldwide. Although it receives considerable attention, the early steps of the host response remain poorly defined. Here, we report a noninvasive method using milk fat globules (MFG) as a source of mammary RNA to follow the dynamics of the global transcriptional response of mammary epithelial cells (MEC) during the course of a bacterial infection. We first assessed that RNA isolated from MFG were representative of MEC RNA; we then evaluated whether MFG RNA could be used to monitor the MEC response to infection. Sufficiently high yields of good-quality RNA (RNA integrity numbers ranging between 6.7 and 8.7) were obtained from goat MFG for subsequent analyses. Contamination of MFG by macrophages and neutrophils, which can be trapped during creaming, was assessed and when using quantitative real-time PCR for cell-type specific markers, was shown to be weak enough (<8%) to affect MFG gene expression profiling. Using microarrays, we showed that RNA extracted from MFG and from mammary alveolar parenchyma shared approximately 90% of the highlighted probes corresponding in particular to genes encoding milk proteins (CSN, BLG, LALBA) and enzymes involved in milk fat synthesis and secretion (FASN, XDH, ADRP, SCD, and DGAT1). In addition, a gene involved in the acute-phase reaction, coding for the serum amyloid A3 (SAA3) protein, was found within the first 50 most highly expressed genes in a noninfectious context in both mammary alveolar parenchyma and MFG, strongly suggesting that SAA3 is expressed in MEC. We took advantage of this noninvasive RNA sampling to follow the early proinflammatory response of MEC during the course of a bacterial infection and showed that the levels of mRNA encoding SAA3 sharply increased at 24h postinfection. Taken together, our results demonstrate that MFG represent a unique source of MEC RNA to noninvasively sample sufficient amounts of high-quality RNA to assess the dynamics of MEC gene expression in vivo, especially during the first steps of infection, thereby paving the way for the discovery of early biomarkers for the control of intramammary infections. Furthermore, this noninvasive technique could be used to provide mammary transcriptomic data on a large scale, thus filling the gap between genomic and phenotypic data.