The mammary gland consists of an organized network of epithelial ducts and lobules. This histoarchitecture can be recapitulated in vitro by culturing mammary epithelial cells as 3D spheroids embedded in a reconstituted basement membrane. I first used this assay to characterize the role of cell-cell and cell-ECM adhesion in the formation and polarization of the apical junction complexes in normal mammary epithelial cells. Cell-cell adhesion alone was sufficient to initiate polarized junction assembly. However, the addition of exogenous ECM generated a spatial polarity signal dependent on laminin-1 and α6 and β1 integrins. This caused clusters of mammary epithelial cells to re-localize the junctional complexes to the center of the spheroid prior to lumen formation. In ductal breast carcinoma, a critical hallmark is the loss of normal polarized tissue architecture without the induction of an epithelial-to-mesenchymal transformation (EMT). Thus, misregulation of molecules that function as polarity determinants may contribute to ductal tumor progression. Podocalyxin is an anti-adhesive glycoprotein that may be involved, as it is important in epithelial morphogenesis, and its overexpression in clinical breast tumors is associated with poor outcome. Despite this, overexpression of podocalyxin in normal mammary epithelial cells did not disrupt 3D morphogenesis or apicobasal polarity. However, its overexpression in non-metastatic breast tumor cells did perturb the architecture and growth of tumor spheroids in vitro and it facilitated subcutaneous tumor growth in vivo without causing an EMT. Mechanistically, podocalyxin localized to and expanded non-adhesive membrane domains and induced microvillus formation that was dependent on its extracellular domain and Rho GTPase-regulated actin polymerization. Podocalyxin also recruited its intracellular binding partners NHERF-1 and ezrin via its cytoplasmic tail. Strikingly, the formation of this protein complex was not required for microvillus formation. Additionally, podocalyxin delayed cell-cell aggregation and decreased the initial adhesion, spreading and strength of attachment of tumor cells to fibronectin where it restricted β1 integrin localization to the basal/attached domain. These alterations in adhesion possibly contributed to podocalyxin's ability to increase growth factor-dependent tumor cell migration. Altogether, these data indicate that podocalyxin overexpression may facilitate a ductal tumor-like progression that involves EMT-independent alterations in tissue architecture.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1991. / Vita. Abstract. Includes bibliographical references (leavea 118-121). Also available via the Internet.

Thesis (M. Sc.)--University of Canterbury, 2009. / Typescript (photocopy). Includes bibliographical references (leaves 57-63). Also available via the World Wide Web.

Thesis (M.S.)--University of Wisconsin--Madison, 1969. / eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (Ph. D.)--University of Washington, 2007. / Vita. Includes bibliographical references (leaves 132-182).

The mammary gland consists of an organized network of epithelial ducts and lobules. This histoarchitecture can be recapitulated in vitro by culturing mammary epithelial cells as 3D spheroids embedded in a reconstituted basement membrane. I first used this assay to characterize the role of cell-cell and cell-ECM adhesion in the formation and polarization of the apical junction complexes in normal mammary epithelial cells. Cell-cell adhesion alone was sufficient to initiate polarized junction assembly. However, the addition of exogenous ECM generated a spatial polarity signal dependent on laminin-1 and α6 and β1 integrins. This caused clusters of mammary epithelial cells to re-localize the junctional complexes to the center of the spheroid prior to lumen formation. In ductal breast carcinoma, a critical hallmark is the loss of normal polarized tissue architecture without the induction of an epithelial-to-mesenchymal transformation (EMT). Thus, misregulation of molecules that function as polarity determinants may contribute to ductal tumor progression. Podocalyxin is an anti-adhesive glycoprotein that may be involved, as it is important in epithelial morphogenesis, and its overexpression in clinical breast tumors is associated with poor outcome. Despite this, overexpression of podocalyxin in normal mammary epithelial cells did not disrupt 3D morphogenesis or apicobasal polarity. However, its overexpression in non-metastatic breast tumor cells did perturb the architecture and growth of tumor spheroids in vitro and it facilitated subcutaneous tumor growth in vivo without causing an EMT. Mechanistically, podocalyxin localized to and expanded non-adhesive membrane domains and induced microvillus formation that was dependent on its extracellular domain and Rho GTPase-regulated actin polymerization. Podocalyxin also recruited its intracellular binding partners NHERF-1 and ezrin via its cytoplasmic tail. Strikingly, the formation of this protein complex was not required for microvillus formation. Additionally, podocalyxin delayed cell-cell aggregation and decreased the initial adhesion, spreading and strength of attachment of tumor cells to fibronectin where it restricted β1 integrin localization to the basal/attached domain. These alterations in adhesion possibly contributed to podocalyxin's ability to increase growth factor-dependent tumor cell migration. Altogether, these data indicate that podocalyxin overexpression may facilitate a ductal tumor-like progression that involves EMT-independent alterations in tissue architecture. / Medicine, Faculty of / Graduate

The porcine mammary gland can be used for the production of recombinant proteins by directing a transgene to the mammary gland with a milk protein gene promoter. In order to determine whether or not the protein will be expressed, the animals must be maintained at least through their first lactation. An experiment was performed to determine if hormonal induction of lactogenesis in transgenic virgin pigs could be used as a method for identifying those gilts that are likely to express the recombinant protein during a natural lactation. Mammary development and lactogenesis were induced by administration of subcutaneous implants designed to release 7.1 mg of estradiol-17 beta and 18 mg of progesterone daily for 21 d. Histological analysis of tissue samples before and after the treatment period indicated that mammary secretory tissue underwent dramatic proliferation resulting in a greater degree of alveolar and individual epithelial cell differentiation. The presence of beta-lactoglobulin mRNA was detected in high levels in post-implant tissue samples, and minimally detected in samples cultured in media supplemented with insulin, hydrocortisone, and prolactin. However, protein expression was only detected in the post-implant samples, indicating that beta-lactoglobulin was not maintained well by in vitro culture. The transgene mRNA, recombinant human fibrinogen (A-alpha chain), was detected in all analyzed samples at varying levels. However, the corresponding protein was not detected in any sample, under either reduced or nonreduced conditions. These results indicate that lactogenesis was successfully induced using the hormonal implants. Also, the transgene was activated by the hormonal induction in vivo and in vitro, but the corresponding protein could not be detected. This study indicates that induction of lactogenesis can be used to detect the presence of transgene mRNA in mammary tissue of gilts. However, we cannot conclusively demonstrate that this procedure can be used to identify those gilts that are likely to express the recombinant protein during a natural lactation. / Master of Science

No description available.

No description available.

PEA3 is the founding member of a subfamily of closely related Ets transcriptional regulatory proteins that includes ERM and ER81. The PEA3 subfamily members share greater than 95% identity in their ETS DNA binding domain and 500/o sequence similarity overall, suggesting these genes may serve redundant functions. The overexpression of each member is positively correlated with HER2 mediated breast tumorigenesis in humans and mice, suggesting a role for this subfamily in mammary development and oncogenesis. This study first addresses the role of PEA3 in cellular transformation mediated by oncogenic Ras and Neu. Wildtype and PEA3-null mouse embryo fibroblast cell lines were infected and tested for focus formation. PEA3-null fibroblasts are refractory to transformation as compared to their wildtype counterparts. Ras and Neu transformed foci show elevated PEA3 subfamily mRNA transcripts and PEA3 protein. ERM and ER81 are expressed in PEA3-null fibroblasts and do not appear to compensate for loss of function mutations in the PEA3 gene resulting in the transformation-defective phenotype. Expression of candidate PEA3 target genes (MMP-3 and MMP-9, which have known roles in transformation) is compromised in PEA3-null fibroblasts. Re-expression of PEA3 in these cells rescues the transformation-deficient phenotype and restores expression of MMP-3 and MMP-9. Hence, PEA3 appears to be a crucial effector in Ras and Neu mediated transformation, in addition to serving an important regulatory role of genes involved in cell motility and invasive tumor behaviour. This study also addresses the role of ER81 in normal mammary gland development. PEA3 is required for normal mammary gland development, as displayed by the reduced branching phenotype in PEA3-null female mice. Mice lacking functional ER81 were generated to determine if ER81 serves a similar role in mammary gland development. ER81 is expressed in the epithelial cells of mammary buds at E 1 0.5, when these structures first appear during mouse embryogenesis. ER81 is then differentially expressed during postnatal mammary gland development, with highest expression occurring at times of extensive epithelial branching. During puberty, expression is observed in undifferentiated cap and body cells of terminal end buds, in differentiated luminal and myoepithelial cells of ducts. During pregnancy, expression in luminal epithelial cells is lost, but persists in the myoepithelial cells within the ducts and alveoli. Targeted disruption of both ERSt alleles result in severely runted mice that die by 4 weeks of age, thereby precluding study of mammary gland development in these mice beyond this developmental stage. However, loss of a single ER81 allele results in healthy looking mice, comparable in size and lifespan to wildtype littermates. Studies employing ER81 heterozygous mice reveal a 50% allelic dose is sufficient for normal mammary gland development. Loss of a single ER81 allele did not result in any overt phenotypes in ductal branching, lobulo-alveolar development, or morphology of the surrounding fat pad. / Thesis / Master of Science (MSc)