Despite the increasing evidence that some breast tumours might arise from a tumour stem cell, current transgenic mouse models of this cancer generally use gene promoters that are active primarily in differentiated cells. An innovative model of breast cancer developed by Alana Welm and colleagues redresses this and provides evidence that MET and MYC, when overexpressed in progenitor cells, cooperate in breast tumorigenesis.

The authors used a retroviral vector called pMIG, which was developed from mouse stem-cell virus, with the aim that it would be expressed in progenitor or stem cells. This was confirmed by using the vector to express green-fluorescent protein (GFP) in mouse primary mammary epithelial cells, and transplanting the cells into mouse fat pads, which forced the cells to expand. Moreover, the cells developed into GFP-expressing mammary glands.

Overexpression of the proto-oncogene MET is evident in a subset of breast cancers and is strongly associated with poor prognosis, but little is known about its involvement in tumorigenesis. Also, MET expression increases fivefold during the differentiation of human breast progenitor cells. So Welm and colleagues isolated mammary epithelial cells from a transgenic mouse that expresses MET under the control of a tetracycline-response element. These cells were infected with pMIG expressing a tetracycline-repressible transactivator protein. Once the cells were transplanted, the recipient mice were either treated with the tetracycline derivative doxycycline, to repress MET expression, or left untreated, for overexpression driven by the transactivator.

Overexpression of MET led to the development of neoplastic clusters of epithelial cells in the reconstituted mammary glands, unlike the doxycycline-treated glands. Immunohisto-chemistry confirmed that the MET protein was actively signalling in the abnormal cells. Importantly, the overexpression of MET correlated with presence of the CK6 protein, a candidate marker for breast progenitor cells. However, the mice did not develop malignant tumours, indicating the involvement of other factors. The authors theorized that one of these might be MYC, which is also overexpressed in some breast cancers, although whether MET and MYC are usually co-expressed in human tumours is not known. Using the pMIG vector to overexpress MYC and MET in the mammary cells produced focal palpable tumours in the reconstituted glands; no tumours occurred in glands expressing MYC alone. Although the MYC/MET tumour cells were highly proliferative, no metastases were found.

The authors suggest that their approach will be useful in examining how other genes affect mammary tumorigenesis when expressed in progenitor cells. In addition, the use of this model might help to determine the characteristics of breast cancers that arise from less differentiated mammary epithelial cells.