Mammary epithelial cells are embedded in a unique extracellular environment to which adipocytes and other stromal cells contribute. Mammary epithelial cells are critically dependent on this milieu for survival. However, it remains unknown which adipocyte-secreted factors are required for the survival of the mammary epithelia and what role these adipokines play in the process of ductal carcinoma tumorigenesis. Here, we take a systematic molecular approach to investigate the multiple ways adipocytes and adipokines can uniquely influence the characteristics and phenotypic behavior of malignant breast ductal epithelial cells. Microarray analysis and luciferase reporter assays indicate that adipokines specifically induce several transcriptional programs involved in promoting tumorigenesis, including increased cell proliferation (IGF2, FOS, JUN, cyclin D1), invasive potential (MMP1, ATF3), survival (A20, NFjB), and angiogenesis. One of the key changes in the transformed ductal epithelial cells associated with the cell cycle involves the induction of NFjB (five-fold) and cyclin D1 (three-fold). We show that by regulating the transcription of these molecules, the synergistic activity of adipocyte-derived factors can potentiate MCF-7 cell proliferation. Furthermore, compared to other stromal cell-secreted factors, the full complement of adipokines shows an unparalleled ability to promote increased cell motility, migration, and the capacity for angiogenesis. Adipocyte-secreted factors can affect tumorigenesis by increasing the stabilization of pro-oncogenic factors such as b-catenin and CDK6 as a result of a reduction in the gene expression of their inhibitors (i.e. p18). An in vivo coinjection system using 3T3-L1 adipocytes and SUM159PT cells effectively recapitulates the host-tumor interactions in primary tumors. Type VI collagen, a soluble extracellular matrix protein abundantly expressed in adipocytes, is further upregulated in adipocytes during tumorigenesis. It promotes GSK3b phosphorylation, b-catenin stabilization, and increased b-catenin activity in breast cancer cells and may critically contribute towards tumorigenesis when not counterbalanced by other factors.

Active cell death (ACD) in hormone-dependent tissues such as the prostate and mammary gland is readily induced by hormone ablation and by treatment with anti-androgens or anti-estrogens, calcium channel agonists and TGF beta. These agents induce a variety of genes within the hormone-dependent epithelial cells including TRPM-2, transglutaminase, poly(ADP-ribose) polymerase, Hsp27 and several other unidentified genes. Not all epithelial cells in the glands are equally sensitive to the induction of ACD. In the prostate, the secretory epithelial cells that are sensitive to hormone ablation are localized in the distal region of the prostatic ducts, and are in direct contact with the neighboring stroma. In contrast, the epithelial cells in the proximal regions of the ducts are more resistant to hormone ablation, probably because the permissive effects of the stroma are attenuated by the presence of the basal epithelial cells, which are intercalated between the epithelium and stroma. The underlying biology of ACD in prostate and mammary glands, and its relevance to hormone resistance, is discussed in this review.

No abstract

Poly (A)+ RNA from the prostates of both intact and castrated rats was translated in a message-dependent reticulocyte lysate, and the translation products were electrophoresed on SDS/polyacrylamide gels. Fluorography of these gels showed the expected disappearance, after castration, of the prostate steroid-binding proteins as well as a number of other androgen-dependent proteins. Two major (Mr 40,000 and 45,000) and several minor proteins appeared in the translation products of the castrated rat prostate RNA. Criss-cross liquid hybridization analysis between prostate poly (A+) RNA from intact and castrated rats also showed the disappearance of the abundant prostate steroid-binding protein sequences after castration and the synthesis of several new low to medium abundance sequences. Northern hybridization experiments demonstrated the presence of at least two, and possibly four androgen-repressed poly (A)+ RNA sequences. The most prominent of these, an RNA of 2,000 nucleotides, appeared within 2 days of castration, reaching a maximum on day 4 at a level approximately 400 times greater than the normal level. The other major sequence (a sequence of 1,000 nucleotides) appears after 4 days, reaching a peak between days 8 and 11. Sequences similar to these new RNAs could play an important role in the long-term resistance of prostatic cancer to hormone therapy in humans.

In LNCaP prostate cancer cells CG-1521, a new inhibitor of histone deacetylases, alters the acetylation of p53 in a sitespecific manner. While p53 is constitutively acetylated at Lys320 in LNCaP cells, treatment with CG-1521, stabilizes the acetylation of p53 at Lys373, elevating p21 (and inducing cell cycle arrest). Treatment with CG-1521 also promotes Bax translocation to the mitochondria and cleavage, and apoptosis. TSA stabilizes the acetylation of p53 at Lys382, elevating p21 levels and inducing cell cycle arrest, but does not induce Bax translocation or apoptosis. In LNCaP cells CG-1521, but not TSA, promotes the rapid degradation of HDAC2. These data suggest that the acetylation of p53 at Lys373 is required for the p53-mediated induction of cell cycle arrest and apoptosis, while acetylation of p53 at Lys382 induces only cell cycle arrest. In p53 À/À PC3 cells both compounds induce p21 and cell cycle arrest, but not Bax translocation or apoptosis, suggesting that both compounds can also induce p21 through a p53-independent mechanism.

1,25-dihydroxyvitamin D3 [1,25(OH)2D3], the active metabolite of vitamin D2 inhibits breast cancer cell growth both in vivo and in vitro. In addition to its anti-proliferative effects, 1,25(OH)2D3 induces morphological and biochemical markers of apoptosis in MCF-7 cells. In the studies reported here, we compared the effects of 1,25(OH)2D3 and EB1089, a low calcemic vitamin D analog, on cell cycle kinetics and apoptosis in MCF-7 cells. Both vitamin D compounds reduced viable MCF-7 cell number in a time and dose dependent manner, with EB1089 approximately 50 fold more potent than 1,25(OH)2D3. Flow cytometric analysis indicated that both agents induced cell cycle arrest in G, G1 which was associated with accumulation of the hypophosphorylated form of the retinoblastoma (Rb) protein. MCF-7 cells treated with either 1,25(OH)2D3 or EB1089 for 48 h exhibited characteristics of apoptosis, including cytoplasmic condensation, pyknotic nuclei, condensed chromatin and DNA fragmentation. Cells treated with either agent exhibited up regulation of proteins associated with mammary gland regression (clusterin and cathepsin B) and down regulation of the anti-apoptotic protein bcl-2. These studies demonstrate that, despite its lower calcemic activity in vivo, the vitamin D analog EB1089 induces effects that are indistinguishable from those of 1,25(OH)2D3 on cell number, cell cycle and indices of apoptosis in MCF-7 cells in vitro. In addition, since both agents rapidly down regulate estrogen receptor, disruption of estrogen dependent signalling may play a role in the induction of apoptosis by vitamin D compounds in MCF-7 cells.

The TRPM‐2/clusterin gene and its cognate protein has been characterized in a number of species. Although the functional role, or roles, of the TRPM‐2/clusterin protein remains to be firmly established, the gene has been implicated in a variety of physiological processes, including sperm maturation, lipid transport, membrane remodelling and inhibition of the complement cascade. TRPM‐2/clusterin is induced de novo during the regression of the prostate and other hormone‐dependent tissues after hormone ablation, and is over‐expressed in several human neurodegenerative diseases including Alzheimer's disease, epilepsy and retinitis pigmentosa. We describe the genomic structure of the human TRPM‐2/clusterin gene which is organized into nine exons, ranging in size from 47 bp (exon I) to 412 bp (exon V), spanning a region of 16580 bp. Comparison with sequences registered in the databases shows that it has extensive similarity to the human protein designated as SP‐40,40 or complement‐lysis inhibitor (CLI), a protein that appears to block the membrane‐attack complex of complement. However, the cDNA sequences reported for SP‐40,40 and CLI diverge significantly in the 5′ untranslated region of the mRNA (coded for by exon I), raising the possibility that the TRPM‐2/clusterin gene is present in the human genome as a small multi‐gene family or that there are several alternate exon I sequences in the TRPM‐2 gene. Southern analysis and fluorescent in situ hybridization suggest that the clusterin gene is a single‐copy gene, and that, if alternative exon I sequences are present in the genome, they lie outside of the λ clones that have been characterized. Analysis of the promoter region of the human TRPM‐2/clusterin gene shows many similarities to the rat TRPM‐2/clusterin promoter including a putative control region containing several potential regulatory elements that may regulate the complex tissue‐specific control of the gene which must be constitutively expressed in some tissues but is inducible in others.

No abstract