For many years it has been tacitly assumed that actions of estradiol and on epithelium are mediated directly through ER in epithelial cells themselves. Mammary ductal growth gland is induced by estradiol. The presence of ERα in mammary epithelium combined with the ability of estradiol to induce proliferation in mammary epithelium certainly suggests a causal link; i.e., that effects of estradiol on mammary epithelial proliferation are mediated directly through ER in these cells. However, such a causal link between the presence of mammary epithelial ER and the effects of estradiol on proliferation of mammary epithelium has never been established. In fact, a rapidly emerging body of evidence suggests that this is clearly not true and that  stromal  ER mediates  many effects  of estradiol  on estrogen target epithelium.

       Methodologies for producing αERKO mice, in which the ERα gene has been rendered non-functional by gene targeting, has allowed examination of phenotypic and functional consequences of an absence of ERα. We have recently developed a new experimental system, which utilizes tissues from ERKO mice to study the mechanism of estradiol action. This system involves separating and recombining epithelial and stromal tissue from αERKO mice with that of a wild-type mice, which express ERα. This tissue separation/recombination technique provides a unique method for experimentally controlling the ERα status of both stroma and epithelium. Thus, tissue recombinations can be prepared, which lack ERα in both stromal and epithelial compartments, express ERα in either epithelium or stroma, or express ERα in both epithelium and stroma. Tissue recombinants are then transplanted into host animals. By analyzing effects of a lack of stromal and/or epithelial ERα on estradiol response such as ductal growth, the role of ERα in each tissue compartment can be definitively determined.

2.  STROMAL ERαIS REQUIRED FORDUCTAL GROWTH IN THE MOUSE MAMMARY GLAND

Ductal mammary growth is profoundly impaired in αERKO mice. By the end of puberty, the mammary fat pad (FP) was almost entirely filled with ducts in wild-type mice, while in αERKO mice most of the FP was devoid of 4 ducts .   Analysis   of tissue  recombinants   were  prepared  with  mammary epithelium (MGE) and mammary fat pad (FP) of wild-type (wt) and αERKO mice demonstrated that tissue recombinants composed of wt-FP+wt-MGE developed an extensively branched ductal network, which completely filled the fat pad. In contrast, ductal growth was minimal in αERKO-FP+αERKO-MGE tissue recombinants. When αERKO mammary epithelium was grown in association with wild-type FP (wt-FP+αERKO-MGE), αERKO mammary epithelium underwent extensive ductal growth.    Although there was some variability in the amount of ductal growth in individual tissue recombinants, ductal  growth  was  always  more  extensive  in  wt-FP+wt-MGE  and wt-FP+αERKO-MGE versus αERKO-FP+αERKO-MGE tissue recombinants. Surprisingly,   ductal   tissue   was   never   recognized   in   wholemounts   of αERKO-FP+wt-MGE tissue recombinants. However, serial sections of αERKO-FP+wt-MGE tissue recombinants revealed that all contained small 4 foci of mammary ducts.    These findings demonstrate that stromal fat pad cells are critical estrogen targets, and that estrogen elicits mammary ductal growth  through   stromal   ERa.   Epithelial   ERa   is  neither  necessary  nor sufficient for ductal growth. These findings in mammary gland are in complete agreement with comparable tissue recombinant studies in the uterus and vagina.

          Stromal epidermal growth factor receptor plays a critical role in mammary epithelial growth. Growth factors such as epidermal growth factor are known to play a role in stromal-epithelial interactions which are critical in determining patterns of mammary growth, development, and ductal morphogenesis. To determine the role of signaling through the epidermal growth factor receptor (EGFR) in mammary ductal growth and branching, mice with a targeted null mutation in the EGFR were used. Such EGFR-KO mice die perinatally, and thus transplantation methods were used to study growth and development of EGFR-KO mammary glands. When transplanted under renal capsules of athymic female mice, neonatal mammary glands of wild-type mice underwent extensive ductal growth with complete filling of the FP. Conversely, neonatal mammary glands of EGFR-KO mice exhibited impaired ductal growth with incomplete filling of the FP. These findings demonstrated that EGFR is essential for mammary ductal growth. To determine whether impaired ductal growth was due to an absence of EGFR in epithelium (E), FP or both, EGFR-KO or wild-type MGE was transplanted into wild-type gland-free FPs. Surprisingly, transplants of either EGFR-KO or wild-type MGE into wild-type mammary FPs exhibited extensive ductal growth with comparable complete filling of the FP in transplants into intact virgin female hosts. Similarly, lobulo-alveolar development was equivalent in transplants of EGFR-KO or wild-type MGE into wild-type mammary FPs, when the hosts also received a pituitary isograft as a source of prolactin. These findings suggested that the absence of EGFR-signaling in the epithelium is not required for mammary ductal growth and lobulo-alveolar development. Instead, these findings suggested that the impaired ductal growth exhibited by EGFR-KO mammary glands was due to an absence of EGFR signaling in the mammary FP.

3.   CONCLUSION

         Mammary ductal growth is estrogen-dependent and is profoundly impaired in αERKO mice. Tissue recombinant studies using FPs and mammary epithelia from aERKO and wt mice demonstrate that estrogen stimulates mammary ductal growth via a paracrine mechanism, acting through stromal ERα. Epithelial ERα is neither necessary nor sufficient for ductal growth. EGFR signaling is thought to be a downstream effector of estrogen action in several estrogen target organs. Mammary epithelium expresses EGFR, and epidermal growth factor and TGFα are mitogens for mammary epithelial cells. Despite this, our data clearly show that ductal growth does not require signaling through epithelial EGFR in the mammary gland in vivo. Instead, EGFR is absolutely necessary in the stromal FP to induce estrogen-dependent ductal growth. Our results suggest that under stimulatory estrogenic conditions the stroma responds to estrogen action through an EGFR-mediated signaling event that is required for stimulation ofepithelial growth and development. In contrast, epithelial EGFR and ERa are neither necessary nor sufficient for ductal growth. What are the gene targets of estrogen action in mammary epithelial cells? One target of estrogen action in mammary epithelium is upregulation of the progesterone receptor (PR), which is only detectable in mammary epithelium . The importance of epithelial PR in the mammary gland is demonstrated as impaired lobulo-alveolar development in PR null mice. The targets of EGFR signaling remain to be determined.