Our discoveries of the regulation of CLIC4 nuclear translocation and its function in the TGFbeta pathway have inspired further examination of its role in cancer pathogenesis. Previously we have shown that CLIC4 protein expression is reduced in epithelial cancer cells, and we have now confirmed that this is transcriptionally regulated. The reduction in CLIC4 expression is detected at the benign tumor stage in skin carcinogenesis, with benign tumors at high risk for progression to cancer having the lowest level of expression. Within the last 400 bases of the CLIC4 promoter and into the first exon there is a high frequency of CpG islands. However methylation specific PCR (MS PCR) and combined bisulfite modification and restriction analysis (COBRA) did not reveal promoter methylation. Thus the regulation of CLIC4 expression is cancer cells remains to be illuminated. Nevertheless, apoptosis is induced and tumor growth is inhibited by targeting exogenous CLIC4 to the nucleus of cancer cells. Thus CLIC4 remains a potential therapeutic target and its reduction in cancer cells could underlie the widespread resistance of cancer cells to TGFbeta. Poorly healing wounds increase the risk of skin cancer and the heparan sulfate proteoglycan syndecan1 (sdc1)is essential for normal wound healing. In collaboration with Prof. Mary Ann Stepp we have tested susceptibility of sdc1 null mice to skin tumor formation. In the classic two stage skin painting model, tumor yield was reduced by 50% in the null mice and fewer cancers developed. In contrast a model of cell grafting of oncogenic ras transformed keratinocytes to a skin wound site indicated enhanced tumor formation in cells lacking sdc1. Furthermore, transplanting sdc1 null fibroblasts in the wound bed enhanced tumor formation. Together these results suggest that sdc1 has anti-tumor properties in a wound environment. Cutaneous inflammation is highly associated with skin cancer development. We have developed a mouse model for inducible skin inflammation and have challenged the model in vivo and in vitro to test the role of inflammation in carcinogenesis. The mice are exquisitely sensitive to tumor formation and previous results indicate that tumor sensitivity and inflammation are dependent on upregulation of chemokines MIP-2, KC and GCP2 and activation of their receptor on neutrophils and keratinocytes. Treatment of ras-transformed keratinocytes with exogenous MIP-2 promotes their migration in vitro. That response is absent in CXCR2 deficient cells but can be restored by transducing those cells with an adenovirus encoding murine GFP-CXCR2. Treatment of CXCR2 transduced primary keratinocytes with MIP-2 causes an increase in intracellular calcium concentration. CXCR2 activation leads to ERK and Akt (Ser473) phosphorylation. Activation of ERK by CXCR2 is maintained in EGFR deficient cells excluding the possibility that EGFR transactivation by CXCR2 mediates ERK activation. In collaboration with Victor Marquez and Thomas Craig and David Maloney from NHGRI, we are developing small molecule inhibitors of CXCR2 for treatment of skin tumors and skin inflammation. We have achieved potencies in the 200nM range for in vitro inhibition of CXCR2 activity on keratinocytes and are currently testing these compounds for activity in vivo. The discovery of the participation of CXCR2 and its ligands in cutaneous cancer and inflammation has extended to other model systems. In collaboration with Jeffrey Arbeit we have shown that the exaggerated cutaneous inflammation developing as a consequence of HIF1alpha activity in the mouse epidermis results from cell autonomous activity of TNFalpha and upregulation of CXCR2 ligands. Neutralizing antibodies to these ligands abrogates the HIF1alpha mediated inflammatory response. These mice are now being challenged for tumor formation. Dr. Giorgio Trinchieris laboratory has shown that MyD88 (Myeloid differentiation primary response gene (88)) deficient mice are resistant to two stage skin carcinogenesis. In collaboration we have shown that MyD88 deficiency strongly reduces pro-inflammatory genes (CXCR2 ligands, TNFalpha, GM-CSF) and matrix remodeling (MMP9) in ras transformed keratinocytes. However induction of EGFR ligands (HB-EGF, betacellulin) was not altered. This result suggests that MyD88 deficiency specifically prevents the upregulation of a subset of genes that are normally induced by ras transformation. This concept is now being tested in our orthotopic tumor grafting model. We have reported that the proinflammatory proteins S100A7 and S100A15 are chemokines for specific leukocyte populations, are differentially expressed in skin, are upregulated in skin inflammation and cancer and activate distinct receptors. Their chemokine activities are synergistic when the proteins are coexpressed. We have now targeted the single mouse homologue (mS100A7/15) to epidermis in a tetracycline regulated bigenic system. The transgene causes constitutive infiltration of granulocytes and lymphocytes into the dermis and epidermis and sensitizes the skin of these mice to further exposures to inflammatory stimuli. The mice are also preconditioned for induction of a number of proinflammatory cytokines and chemokines. The action of mS100A7/15 is regulated through the receptor RAGE. This model is now under study for sensitivity to skin cancer induction. We are also interested in potential therapeutic approaches to the most common of all human cancers. PEP005 (ingenol 3-angelate), extracted from Euphorbia peplus, is in clinical trials for treating cutaneous basal cell carcinoma, actinic keratosis, and squamous cell carcinoma in situ by topical application. In mouse models we have found that topical PEP005 induced a rapid inflammatory response, accompanied by hemorrhage when applied to the flank of athymic nude mice.The closely related but not therapeutic PMA induced similar inflammation but no hemorrhage, and the consequences of these two agents on mouse skin and subcutaneous tumors were compared.Three consecutive daily treatments of PEP005 eliminates or reduces SCC and murine melanoma xenografts whereas PMA does not. Application of cyclosporine A before each treatment of PEP005 reduces the anti-cancer activity in vivo.In contrast,eliminating neutrophil infiltration by dexamethasone has no effect on PEP005 induced tumor ablation. Tissue sections reveal that the blood vessels in the normal tissue surrounding tumor and in the periphery of tumors are damaged one day after the initial PEP treatment, and blood clots are formed in those blood vessels, leading to tumor necrosis. Both PEP005 and PMA activate endothelial cells in vitro and disrupt tube formation that cannot be inhibited by cyclosporine A. This suggests that the cyclosporine A effect in vivo is mediated through a non-enothelial cell type, presumably keratinocytes, that facilitate PEP005 infiltration to blood vessels in the deeper dermis. PEP005, not PMA, blocks [125I] IAAP's binding to ABCB1, the multi drug resistance protein1 (mdr1) in an in vitro photo labeling assay, suggesting PEP005 is a substrate/inhibitor for mdr1. Furthermore, three structurally unrelated mdr1 inhibitors, cyclosporine A, verapamil, and XR9576 all prevent PEP005 induced hemorrhages in the skin of nude mice. We have detec [summary truncated at 7800 characters]
Showing the most recent 10 out of 18 publications
