The expression and nuclear localization of CLIC4 are greatly reduced in human tumor cells. We have generated mice with a tetracycline inducible CLIC4 transgene targeted to the mouse epidermis. CLIC4 is progressively reduced as mouse skin tumor cells evolve through benign and malignant stages, suggesting the mouse skin tumor model will be useful and valid to simulate changes found in human tumors. The tumor microenvironment is increasingly recognized as a major determinant of tumor behavior. Tumor stroma secretes a number of growth and anti-apoptotic factors, supports a variety of inflammatory cells and incorporates new blood vessels. CLIC4 is suppressed in tumor cells in vivo and in vitro but is highly upregulated in tumor stromal cells where it is coexpressed with alpha smooth muscle actin. This myofibroblast transition is also seen in vitro in co-culture experiments where CLIC4 and alpha smooth muscle actin are upregulated in fibroblasts that surround the tumor nests. Furthermore, overexpression of CLIC4 in fibroblasts induces the expression of alpha smooth muscle actin. Fibroblasts treated with TGFbeta in vitro upregulate CLIC4, Schnurri-2 and alpha smooth muscle actin suggesting that the crosstalk seen in human tumors in which CLIC4 is downregulated in tumors and upregulated in stroma could be mediated by secreted TGFbeta. Together these data suggest that CLIC4 participates in the myofibroblast transition of tumor stroma that often is associated with a more aggressive tumor behavior. Inflammation often sensitizes tissues to tumor formation, and it is now increasingly clear that inflammation can contribute to established tumor behavior. We have developed a mouse model for inducible skin inflammation and have challenged the system in vivo and in vitro to examine how inflammation can influence tumor formation. By targeting protein kinase C alpha (PKCalpha) to the epidermis with the keratin 5 promoter, mice develop a marked intraepidermal inflammation upon topical application of an activator of PKC. These mice are exquisitely sensitive to tumor formation in the classical two stage model. The inflammation is mediated through keratinocyte secretion of two cytokines, MIP2 and KC that are the mouse homologues of IL-8, and their interaction with their receptor on neutrophils, CXCR2. Keratinocytes express CXCR2 and could have an autocrine response to the secreted ligands. Ablation of keratinocyte CXCR2 inhibits tumor formation. In vitro, MIP-2 promotes the migration of neoplastic keratinocytes in a CXCR2 dependent manner but does not stimulate proliferation. CXCR2 activation or inhibition does not alter the response of neoplastic keratinocytes to EGFR ligands. However, both Ras activation and EGFR activation increase the expression and secretion of CXCR2 ligands. These changes are through a common EGFR pathway as ablation of EGFR prevents the upregulation of CXCR2 ligands by ras transformation. Together these findings suggest that CXCR2 contributes to cutaneous cancer through an autocrine mechanism and could be a target for anticancer therapy. Inflammation is a poor prognostic sign in several human cancers including breast cancer. Among the sets of chemokines associated with cancer, the S100 proteins are particularly frequent. We have shown the S100A7 and S100 A15 are secreted and chemotactic. Human S100A7 (hS100A7, psoriasin) is associated with tumor progression in breast cancer. hS100A15 is highly homologous to hS100A7 (93% identity), and both paralogs are biologically distinct as shown in our studies. Using specific antibodies generated in our laboratory, we demonstrate that both hS100A7 and hS100A15 are expressed and distributed in different cell types of normal breast tissue and are further differentially upregulated in ductal breast carcinomas. We characterized antibodies that have been widely used to study hS100A7 in breast cancers and other diseases and demonstrate their cross-reactivity with hS100A15. The differential expression and distribution of hS100A7 and hS100A15 using our specific antibodies suggest that both proteins have distinct functions in normal breast and breast tumors. While we continue to explore basic mechanisms of carcinogenesis using in vitro and in vivo models, our attention has also turned towards understanding the mechanism of action of anticancer drugs. PEP005 (ingenol 3-angelate), extracted from Euphorbia peplus, has been used as a traditional medicine for treating warts, corns, skin cancer and other skin conditions. In vitro, PEP005 inhibits cell proliferation and induces apoptosis in a number of human tumor cell lines, including cells from leukemia, colon and lung cancers. Currently, it is in clinical trials for treating basal cell carcinoma, actinic keratosis, and squamous cell carcinoma in situ. Topical treatment of PEP005 eliminates a number of subcutaneously xenografted mouse and human tumors. Topical treatment with PEP005 in three consecutive days temporarily stopped PAM212 tumor growth in a nude mouse xenograft, but caused inflammation accompanied by hemorrhage. Since the tumor and surrounding skin respond similarly to PEP005, we focused our studies on topical treatment of back skin of the athymic nude mice. PEP005-induced inflammation on mouse skin was observed as early as 3 hours after treatment. Infiltration of neutrophils and extravasation of red blood cells (RBC) were observed in the PEP005-treated skin. In contrast, PMA, a classical PKC activator, induced inflammation without hemorrhage and did not cause tumor regression. Electron microscopy revealed that PEP005 treatment damaged blood vessel, including disruption of basement membrane, activation and necrosis of endothelial cells, hemolysis of red blood cells both within and outside of vessels, and altered collagen matrix. Pretreatment with cyclosporin A, FK506, and rapamycin prevented PEP005 induced hemorrhage but not neutrophil infiltration. Our results suggested that PEP005 activates an inflammatory pathway and produces toxicity to blood vessels that could be related to its anti-tumor action and blocked by immunophilin targeting agents. The EGFR has become a major target for therapy of several common human tumors including squamous cell cancer and a number of drugs are in clinical use to inhibit EGFR function. A major dose limiting side effect of anti-EGFR therapy is a troubling folliculitis that has also been reproduced in mice where the EGFR is genetically ablated. Little is known about the changes in skin or tumor homeostasis that may account for the therapeutic or adverse effects of the inhibitors. We have used microarray analysis to address the changes in gene expression that may underlie the therapeutic or off target effects in skin keratinocytes. In order to model the anti-EGFR effect in normal and neoplastic mouse keratinocytes, three pharmacological inhibitors of EGFR were used to inhibit the receptor in wildtype keratinocytes transformed by oncogenic ras, and these cells were compared to inhibitor treated or untreated ras transformed keratinocytes genetically deleted of EGFR. Based on microarray data and further ontology analysis 25 genes have been identified that are differentially regulated when EGFR is absent either genetically or in response to the inhibitors Tarceva, AG1478 and PD153035. In these 25 genes there is a strong enrichment for transporter activity, cell proliferation and transcription factor complex. Microarray data also identified 83 common genes that are off-target changes for the three inhibitors. Of these genes there is a strong enrichment score for immune response and lymphocyte activation. The microarray data have been confirmed with Real Time PCR and functional analyses are in progress
Showing the most recent 10 out of 18 publications
