This laboratory investigates genetic pathways that control developmental programs in mammary epithelium, germ cells and hematopoietic cells. Common to these cell types is their need for cytokines to induce proliferation, survival and differentiation. While prolactin (Prl) controls proliferation and differentiation of mammary epithelium, erythropoietin (EPO) is required for hematopiesis. After binding to their respective receptors, Prl and EPO activate the Jak2/Stat5 pathway and induce developmental programs. We have now demonstrated that in addition to Prl, EGF and GH can activate Stat5 in mammary tissue. Based on tissue culture cell experiments performed by many investigators, Stat5 controls cell survival of many different cell types. A putative target for Stat5 is the bcl-x gene, which encodes a cell survival molecule. We are investigating the contribution of the Jak2/Stat5/Bcl-x pathway in organ development and cancer, with an emphasis on mammary tissue. We have deleted the Stat5a gene from the mouse genome and demonstrated its importance in the functional differentiation of mammary lobulo-alveolar structures during pregnancy. However, mammary alveoli, although undifferentiated, formed in the absence of Stat5a. This suggested that in addition to Stat5a Stat5b was required for the establishment of an alveolar compartment. We approached this question using two strategies. First, we analyzed mice from which both Stat5a and 5b had apparently been deleted, and secondly we initiated experiments to generate mice in which the Stat5a/b locus can be deleted in a cell- and time-specific manner. The available Stat5a/b-null mice are infertile and severely anemic and thus not amenable to direct mammary developmental studies. We have now used a tissue transplantation approach to study the roles of Stat5a/b in mammary development. Stat5a/b-null tissue was transplanted into wild type mice and its pregnancy-mediated development was investigated using histological and molecular tools. From these studies it was clear that Stat5a/b are absolutely required for the specification and formation of alveoli. Without Stat5/b mammary epithelium only formed ducts and no alvoli. Since no alveoli form in the absence of Stat5a/b, it is not possible to study the function of Stat5 in the proliferation, differentiation and survival of mammary alveolar epithelium. In order to be able to conduct such studies, we initiated experiments to delete the Stat5a/b locus in a cell- and time-specific manner. We have cloned and characterized and sequenced the 500 kbp mouse locus containing the genes encoding Stats 3, 5a and 5b. As part of this program we have identified and characterized two new genes, LGP1 and LGP2. LGP1 encodes a protein located in the ER and LGP2 encodes a cytoplasmic protein. Currently we are developing a double targeting approach to introduce loxP recombination sites at the two ends of the locus. We have ES cell lines in which the Stat3 and Stat5 genes had been targeted. However, these cells never gave a high chimerism and it is necessary to start all over. In a quest to identify those proteins in mammary tissue that control cell survival we have determined that Bcl-x is expressed highly at distinct stages during pregnancy and at the interphase of lactation and involution. Bcl-x is a cell survival molecule and there is evidence from tissue culture cells to suggest that the corresponding gene is activated by Stat proteins. We further suggested that Bax could be the counterbalance of Bcl-x in controlled cell survival and death. Studying the role of Bcl-x in mammary development through a gene deletion approach is not feasible since mice, which carry two inactive bcl-x alleles die during fetal development. To circumvent this we established mice that carry a bcl-x gene with loxP sites in the promoter upstream region and in the second exon. The presence of the neomycine gene in the promoter deregulated a germ cell-specific promoter, and both females and males were infertile due the loss of germ cells during fetal development. Based on experimental genetics we identified Bcl-x is a controlling germ cell survival factor in the fetal gonad, and Bax as the controlling death factor. We removed specifically the neomycin selection marker, which restored fertility. Subsequently specific Cre transgenes were introduced. We generated mice that express Cre under control of the whey acidic protein (WAP) gene promoter and the mouse mammary tumor virus (MMTV) long terminal repeat (LTR). WAP-Cre mice express Cre specifically in mammary epithelium during pregnancy. MMTV-Cre mice express Cre in many cell types, including mammary epithelium during puberty and pregnancy, hematopoietic cells and in the ovary. Mammary development during pregnancy and functional performance during lactation were not altered upon the deletion of the bcl-x gene from mammary epithelium. However, the loss of Bcl-x resulted in an accelerated apoptosis during involution. The concomitant loss of Bax did not rescue the premature loss of cells during involution. Deletion of the bcl-x gene in erythrocytes resulted in hemolytic anemia and expansion of immature cells. In contrast to the primordial germ cells, the lack of Bax did not alter the Bcl-x null phenotype. Deletion of the bcl-x gene in oocytes, granulosa cells and luteal cells id not alter their survival, but resulted in reduced fertility. Our work demonstrated a cell-specific and temporally-restricted function of Bcl-x and it revealed a cell-specific balance of the Bcl-x / Bax axis. The retinoblastoma gene encodes a tumor suppressor protein (pRb) that has been linked to many cancers, including breast. We studied the role of pRb in mammary gland development and cancer using experimental mouse genetics. Since pRb-null embryos die in utero, we developed techniques to transplant embryonic mammary epithelium into wild type mice and study their developmental fate. We determined that pRb is not required for functional mammary development. We also introduced several oncogenes expressed in mammary epithelium into a pRb-null background. No acceleration of tumorigenesis was observed. This suggests that the two other members of pocket proteins (p107 and p130) can functionally compensate for the loss of pRb. In order to investigate genes that are controlled by the PrlR/Jak2/Stat5 pathway in mammary tissue we generated cDNA microarrays, which carry more than 6000 genes expressed in mammary tissue. We based these arrays on more than 100,000 Est sequences we obtained form normal and neoplastic mammary tissue. These arrays are now being used by us and other investigators in the field of mammary gland biology.
Showing the most recent 10 out of 113 publications
