Successful colonization by a cancer cell of a distant metastatic site requires immune escape in the new microenvironment. TNF signaling has been implicated broadly in the suppression of immune surveillance that prevents colonization at the metastatic site and therefore must be blocked. In this study, we explored how TNF signaling influences the efficiency of liver metastasis by colon and lung carcinoma in mice that are genetically deficient for the TNF receptor TNFR2. We found a marked reduction in liver metastases that correlated with a greatly reduced accumulation at metastatic sites of CD11b

There is increasing evidence supporting DNA virus regulation of the cell adhesion and tumour suppressor protein, E-cadherin. We previously reported that loss of E-cadherin in human papillomavirus (HPV) type 16-infected epidermis is contributed to by the major viral proto-oncogene E6 and is associated with reduced Langerhans cells density, potentially regulating the immune response. The focus of this study is determining how the HPV16 E6 protein mediates E-cadherin repression. We found that the E-cadherin promoter is repressed in cells expressing E6, resulting in fewer E-cadherin transcripts. On exploring the mechanism for this, repression by increased histone deacetylase activity or by increased binding of trans-repressors to the E-cadherin promoter Epal element was discounted. In contrast, DNA methyltransferase (DNMT) activity was increased in E6 expressing cells. Upon inhibiting DNMT activity using 5-Aza-2′-deoxycytidine, E-cadherin transcription was restored in the presence of HPV16 E6. The E-cadherin promoter was not directly methylated, however a mutational analysis showed general promoter repression and reduced binding of the transactivators Sp1 and AML1 and the repressor Slug. Expression of E7 with E6 resulted in a further reduction in surface E-cadherin levels. This is the first report of HPV16 E6-mediated transcriptional repression of this adhesion molecule and tumour suppressor protein.

Persistent infections with certain high-risk human papillomavirus (HPV) types such as 16 and 18 can result in the development of cervical cancer. Neither of the two prophylactic vaccines against HPV16 and 18 that are in current use have any therapeutic efficacy for prevalent HPV infections. Ablative therapy is widely used for the treatment of HPV cervical dysplasia however disease recurrence is a widely recognized problem. Thus there is a continuing need for therapeutic approaches for the treatment of HPV infections. The HPV16 E6 viral oncoprotein represses surface expression of the cellular adhesion molecule, E-cadherin. Reduced E-cadherin expression on HPV-infected keratinocytes is associated with lowered numbers of antigen-presenting Langerhans cells in the infected epidermis, potentially reducing immune surveillance for HPV. Four chemicals reported to up-regulate E-cadherin were screened for their ability to counteract E6 repression of surface E-cadherin. 5-Aza-2'-deoxycytidine (AzaDC), a DNA methyltransferase inhibitor, and Indole-3-carbinol (I3C), reported to increase E-cadherin through a p21(Waf1/Cip1)-dependent mechanism, had low cytotoxicity and increased or restored E-cadherin expression and adhesive function in HPV16 E6 expressing HCT116 cells. Doxorubicin, also known to induce p21(Waf1/Cip1), increased E-cadherin in E6 expressing cells but had some associated cytotoxicity. Tamoxifen, which can restore adhesive function of surface E-cadherin, was ineffective in counteracting E6 repression of E-cadherin. AzaDC and I3C both show potential to restore antigen-presenting cells to HPV infected skin by antagonizing E6 repression of E-cadherin, thereby counteracting an important immune evasion mechanism of HPV16 and reinstating immune function at the infected site.

<p>Supplementary Fig 1. Identification of hepatic cells that express TNFR2. Supplementary Fig 2. Loss of TNFR2 expression does not affect the induction of CAM expression on LSEC. Supplementary Fig 3. Additional characterization of MDSC isolated from wild type and TNFR2-/- mice. Supplementary Fig 4. The recruitment of Treg cells to the liver is not reduced in TNFR1-/- mice. Supplementary Fig 5. TNFR2 expression is not required for tumor or TNF-α - induced upregulation of hepatic CXCR2 chemokines. Supplementary Fig 6. Loss of TNFR2 does not affect the expression of CXCR1 and CXCR2 on MDSC. Supplementary Fig 7. Reduced TNFR2 expression in the livers of mice treated with TNFR2 ASO. Supplementary Fig 8. HLA-DR is widely expressed in the normal liver.</p>

<p>Supplementary Table 1. List of antibodies and conditions. Supplementary Table 2. List of qPCR primers. Supplementary Table 3. Oligonucleotide sequences</p>