Primary mammary epithelial stem cells persist throughout the reproductive lifespan of the animal, allowing the gland to develop and differentiate with each pregnancy. However, serial transplantation studies indicated that mouse mammary epithelial stem cells are not immortal. Mammary epithelium can be transferred serially to host animals up to seven times before it becomes incapable of repopulating the mammary fat pad, although more than 75% of epithelial outgrowths lose their repopulating ability by the fourth passage (46).
Experiments on the repopulating capacity of mammary epithelium isolated from mice of different ages and reproductive histories suggested that loss of capacity to self-renew is dictated not by chronological age, but by the number of times a stem cell divides; it has been calculated that stem cells become senescent after 40-50 divisions (47,48). Interestingly, this figure agrees with the "Hayflick" number, which is the maximum number of divisions a eukaryotic cell can undergo in culture before replicative senescence
Fig. 1. A model of tissue-specific stem cells and their progeny postulated to exist in the mammary epithelium. The putative stem cells occupy an intermediate location in the epithelium in terms of their physical position and specific protein expression and are defined by their appearance (small and pale staining), ability to efflux Hoechst 33342, retention of DNA label, and in mice, expression of stem cell antigen (Sca-1). We speculate that the stem cell pool includes primary, long-lived cells with a genome that is extremely well protected against damage and a second population of more committed stem cells with a shorter lifespan that contain steroid receptors. The progeny of these stem cells may form a transit-amplifying population before commitment to the ductal and lobular lineages and then final differentiation into luminal and myoepithelial cells. We postulate that the genome of the primary stem cell population is very highly protected, whereas that of the more committed steroid receptor-containing stem cells is not. This means that the latter population may be more susceptible to transforming events, leading to loss of normal growth constraints, proliferation in response to signals such as estrogen and progesterone, and then progression to hyperplasia, to ductal carcinoma in situ, and ultimately to invasive breast tumors.
occurs (49). However, it is probable that no stem cell actually undergoes 40-50 divisions under normal circumstances.
In the experiments that showed that just one cell can reconstitute an entire mammary gland, the stem cell was estimated to have undergone 11 self-renewing symmetrical divisions (27). So, it would appear that no one stem cell survives throughout the entire lifespan of the tissue, but that the population turns over slowly by infrequent self-renewing symmetrical division.
It is still not clear whether the stem cell population size expands and contracts in response to the cyclical developmental needs of the mammary gland. Ultrastructural examination of rat mammary epithelium at all developmental stages indicated that, from nulliparity through to involution, the proportion of SLCs remains constant at about 3% of the total epithelium (34). This suggests that the stem cell population increases and decreases at the same rate as the more differentiated epithelial cells. However, given what is known about stem cell properties in other tissues, it might be expected that absolute mammary stem cell numbers would remain constant; therefore, their proportion would decrease during epithelial cell growth and increase during involution.
One explanation for this disparity is that SLCs are a heterogeneous population that includes both primary mammary stem cells and lineage-committed progenitors; it is the latter population that alters in number during mammary gland growth and involution. This is supported by the results of the limiting dilution transplantation studies of Kordon and Smith, which indicated a primary stem cell frequency of approx 1 in 2500 (0.04%) (27). This is not only much lower than the proportion of SLCs, but is also lower than the number of LRCs or SP cells detected in other studies (38,39) and suggests that label retention and Hoechst 33342 efflux are properties common to both primary stem cells and the more lineage-committed progenitors.
Another assumption is that, given that stem cells appear to divide infrequently, it might be expected that all the stem cells present in the adult mammary gland are direct descendants of those established during embryogenesis. However, Wagner and colleagues suggested that a new "adjunct" epithelial population arises during pregnancy, and that this could explain the physiological differences between nulliparous and involuted parous mammary glands in terms of sensitivity to hormones and carcinogenic agents (50). Using genetic techniques to generate mammary epithelial-specific expression of the P-galactosidase gene, Wagner et al. show that this new population of cells survives postlactational involution and tissue remodeling to give rise to new alveolar cells in subsequent pregnancies. Moreover, the new adjunct cells appear to be able to contribute to both ductal and alveolar epithelial cell types when transplanted into cleared mammary fat pads. The authors concluded that the adjunct epithelial population could represent a new stem cell population derived from more differentiated cells; this population contributes to the biological changes occurring in the mammary epi thelium after pregnancy and lactation. As these changes result in a decreased risk of breast cancer if they occur early enough in a woman's reproductive lifespan, it will be important to determine whether such an adjunct population of cells exists in parous human breast and, if so, what their properties are. The conclusions from these studies is that the mammary epithelial stem cell population is not as fixed in either number or characteristics as first thought.
Was this article helpful?