Bovine mammary stem cells: a candidate vector for transgene expression in milk

In the bovine species, milk production has been shown to be strongly correlated to mammary cell numbers. Right after peak lactation, mammary epithelial cells slowly decrease since the apoptotic rate is slightly higher than proliferation. Most advances in optimizing milk production have been achieved by genetic selection through breeding and husbandry practices. A better understanding of the cells that are able to generate mammary tissue may facilitate improvement in milk production and therefore be of important economic relevance. We show that clonogenic progenitors can be found in the bovine mammary tissue: when cultured in suitable conditions these primitive cells generate lineage-restricted colonies that are characterized by the expression of markers specific for the luminal or myoepithelial cell population. Even more interestingly, when dissociated bovine mammary cells were embedded in collagen gels and subsequently transplanted under the kidney capsule of immunodeficient NOD/SCID mice, organized bilayered structured were detected. Not only these outgrowths are morphologically similar to mammary alveoli found in bovine tissue, but they also share the same expression pattern for several proteins, among which cytokeratin 14 (CK14), cytokeratin 18 (CK18), p63, Smooth Muscle Actin (SMA), MUC1 and EpCAM. Moreover this regenerated alveoli proved to be functional as milk proteins were detected in their lumen when recipient mice were made pregnant. When these outgrowths were dissociated and cells cultured at clonal density, the same lineage-restricted colonies generated by the primary mammary cells were detected along with colonies that displayed a different pattern of expression for the CK14/CK18 proteins. Colonies containing both luminal and myoepithelial cells or double positive cells were found. Several surface antigens have been shown to be differentially expressed in both mouse and human mammary epithelial subpopulations. Unfortunately antibodies that recognize their bovine counterpart are scarce. While EpCAM and MUC1 expression has been tested for FACS analysis, the lack of a suitable antibody to mark the basal subpopulation prevents at the moment the adoption of antibody-based cell sorting strategies. However we found that stem cells and myoepithelial progenitors lack Aldehyde Dehydrogenase 1 (ALDH1) activity and therefore can be enriched using a substrate-based staining procedure coupled with cell sorting. This population was then infected with a lentiviral vector carrying the human β-CASEIN gene and subsequently transplanted in recipient mice. Some of the alveoli regenerated in vivo by these cells were shown to produce and secrete in the lumen the transgenic protein. Our findings indicate that the bovine mammary gland is organized in a hierarchy that is similar to the one described in the mouse and human species. Moreover the in vivo production of a transgene by the regenerated progeny of bovine mammary stem cells represents an important and innovative first step towards novel milk engineering strategies, which may prove of relevant commercial interest. This research has been supported by a CRC 2008 grant.