PerspectiveSTEM CELLS

Decoding hormones for a stem cell niche

See allHide authors and affiliations

Science  21 Apr 2017:
Vol. 356, Issue 6335, pp. 250
DOI: 10.1126/science.aan1506

The ability of certain adult stem cell types to differentiate into other cell types, expand massively, or quiesce until needed underlies the study of development and of how cells grow uncontrollably in cancer. Adult stem cell behavior in many organs depends on unique, protected microenvironments harboring specialized extracellular matrix proteins, support cells, hormones, or other factors collectively called stem cell niches (1, 2). But what, exactly, constitutes a stem cell niche? On page 284 of this issue, Zhao et al. (3) report an element of the mammary gland stem cell niche that drives local expansion and growth of the epithelial compartment in response to global hormones. Select fibroblast cells that are peripheral to the stem cells “translate” hormones into signals that stem cells recognize.

Decoding hormone signals

Epithelial and stromal cells secrete factors in response to global hormones such as estrogen and progesterone. Mammary stem cells then respond to these secreted factors.


The adult mammary gland is composed of a network of epithelial ducts and lobules that synthesize and transport milk and is embedded in the stroma of the mammary fat pad. The epithelial portion expands in response to hormones during puberty or lactation, an ability conferred by abundant adult stem cells. These stem cells can differentiate into the two cell types that constitute the duct system—luminal epithelial and myoepithelial cells (1). Although mammotropic hormones, such as progesterone and estrogen, stimulate mammary epithelial growth and expansion, mammary stem cells appear to be insensitive to them (4, 5). Thus, growth of mammary stem cells poses a conundrum. Mature epithelial cells can sense these hormones and secrete factors that stimulate mammary stem cell expansion (4, 5). However, control of epithelial growth does not depend on the epithelia alone; nonepithelial supportive tissue of the mammary gland (stroma) also controls the extent of epithelial expansion and branching (6). But how could stroma control mammary stem cells in response to hormones?

Glioma-associated oncogene family zinc finger 2 (GLI2), a hedgehog pathway transcription factor, appears to link abnormal breast development and tumorigenesis (7) to the stroma of the mammary gland. Zhao et al. report that loss of GLI2 in a mouse model resulted in abnormal, hypoplastic development of epithelial tissue and the loss of epithelial regenerative capability, indicators of aberrant mammary stem cell function.

Surprisingly, Zhao et al. discovered that mouse mammary epithelial cells do not express this transcription factor. Instead, GLI2 was expressed in a subpopulation of stromal fibroblasts that are closely associated with the outside of epithelial ducts, but do not directly contact epithelia. This fibroblast population in vivo responded to growth hormone or estrogen stimulation by secreting multiple growth factors, including insulin-like growth factor 1 (IGF1), hepatocyte growth factor (HGF), and several wingless-related integration site proteins (WNTs), along with extracellular matrix proteins. Among these many factors, IGF1 and WNT2B that were released from a polymer implanted in the mammary gland rescued mammary development in mice lacking GLI2. Collectively, the results suggest that hormone-driven provision of growth factors by stromal fibroblasts is essential to mammary stem cell niche function.

Communication between stroma and epithelia, and the question of how stem cell populations are maintained in adult organs, take on a new urgency when considering cancer development. A common hypothesis holds that cancers arise from relatively immature cell types (8); thus, the number and potency of stem cells in an organ may be linked to the probability of developing cancer in that organ (9). Patients with congenital growth hormone receptor mutations (Laron syndrome) do not develop cancers (10), whereas overabundance of IGF1 or growth hormones is linked to cancer aggressiveness (11), suggesting that stroma–mammary stem cell communication through this mechanism may also occur in humans. Indeed, genetic background of the stroma can alter breast tumor susceptibility independently of the genetic background of the epithelia even though the tumor arises from epithelial cells (6).

That a specific subpopulation of fibroblasts mediates hormone sensing in the breast is thus an exciting mechanism that links stroma to epithelial growth (see the figure). Altering the number of mammary stem cells or their growth and differentiation potential through stroma–stem cell communication may afford a mechanism for chemoprevention in addition to explaining mammary underdevelopment observed in growth hormone deficiencies.

References and Notes

  1. Acknowledgments: C.R. is funded by the U.S. Department of Defense Congressionally Directed Medical Research Program (BC133875).
View Abstract

Stay Connected to Science

Navigate This Article