Charles S. Tannenbaum

The role of the non-ELR-containing CXC chemokines IP-10 and Mig in antitumor activity induced by systemic treatment with IL-12 was examined in mice bearing the murine renal adenocarcinoma RENCA. IL-12 treatment produces a potent antitumor effect that is associated with tumor infiltration by CD8+ T lymphocytes. The regression of tumor is associated with the elevated expression of the IFN-gamma-inducible chemokines IP-10 and Mig within the tumor tissue. IP-10 and Mig have been shown to function as chemoattractants for activated T lymphocytes. In animals treated with rabbit polyclonal Abs specific for IP-10 and for Mig, the IL-12-induced regression of RENCA tumors was partially abrogated. This effect was associated with a dramatic inhibition of T cell infiltration. Thus, it appears that IL-12-dependent, T cell-mediated antitumor activity requires the intermediate expression of IP-10 and Mig to recruit antitumor effector T cells to the tumor site.

Elevation of intracellular cAMP has been associated with the suppression of macrophage activation. The present study has examined the effects of agents that alter intracellular levels of cAMP on LPS-induced macrophage gene expression. Treatment of murine peritoneal macrophages with trace amounts of LPS leads to dramatically enhanced expression of multiple mRNA including the competence genes JE and KC, first observed in platelet-derived growth factor-stimulated fibroblasts, and those encoding the inflammatory monokines IL-1 and TNF. If macrophages are first treated with cholera toxin or dibutyryl cAMP 15 min before stimulation with LPS, the accumulation of mRNA encoding both JE and TNF is strongly suppressed whereas mRNA levels for KC and IL-1 are unaffected. The suppression of JE and TNF mRNA levels is dose dependent, in the range of 10 to 500 microM dibutyryl cAMP; concentrations as high as 1 mM do not affect the expression of either KC or IL-1. When dibutyryl cAMP is added to macrophages after initiation of LPS treatment, suppressive effects are diminished in a time-dependent fashion. Furthermore, dibutyryl cAMP suppresses the LPS-induced transcriptional activation of the TNF gene. Previous work has shown that the LPS-induced expression of JE appears to be mediated by hydrolysis of polyphosphoinositides and involves a post-transcriptional mechanism. Treatment with dibutyryl cAMP suppresses JE expression induced by treatment with phorbol ester and A23187 suggesting that inhibition of gene expression must act at a site other than the initial transmembrane signaling event. Finally, dibutyryl cAMP only marginally affects the constitutive transcription of the JE gene indicating that suppression may involve a post-transcriptional mechanism. These results indicate that expression of genes encoding inducible early proteins and inflammatory monokines are selectively regulated by elevation of intracellular cAMP. Such effects may be pleiotropic in nature involving multiple molecular mechanisms.

Abstract Purpose: Little is known about the association between myeloid-derived suppressor cell (MDSC) subsets and various chemokines in patients with renal cell carcinoma (RCC) or the factors that draw MDSC into tumor parenchyma. Experimental Design: We analyzed polymorphonuclear MDSC (PMN-MDSC), monocytic MDSC (M-MDSC), and immature MDSC (I-MDSC) from the parenchyma and peripheral blood of 48 patients with RCC, isolated at nephrectomy. We analyzed levels of IL1β, IL8, CXCL5, Mip-1α, MCP-1, and Rantes. Furthermore, we performed experiments in a Renca murine model to assess therapeutic synergy between CXCR2 and anti-PD1 and to elucidate the impact of IL1β blockade on MDSC. Results: Parenchymal PMN-MDSC have a positive correlation with IL1β, IL8, CXCL5, and Mip-1α, and I-MDSC correlate with IL8 and CXCL5. Furthermore, peripheral PMN-MDSC correlate with tumor grade. Given that PMN-MDSC express CXCR2 and parenchymal PMN-MDSC correlated with IL8 and CXCL5, we assessed the response of CXCR2 blockade with or without anti-PD1. Combination therapy reduced tumor weight and enhanced CD4+ and CD8+ T-cell infiltration. In addition, anti-IL1β decreased PMN-MDSC and M-MDSC in the periphery, PMN-MDSC in the tumor, and peripheral CXCL5 and KC. Anti-IL1β also delayed tumor growth. Conclusions: Parenchymal PMN-MDSC have a positive correlation with IL1β, IL8, CXCL5, and Mip-1α, suggesting they may attract PMN-MDSC into the tumor. Peripheral PMN-MDSC correlate with tumor grade, suggesting prognostic significance. Anti-CXCR2 and anti-PD1 synergized to reduce tumor weight and enhanced CD4+ and CD8+ T-cell infiltration in a Renca murine model, suggesting that CXCR2+ PMN-MDSC are important in reducing activity of anti-PD1 antibody. Finally, anti-IL1β decreases MDSC and delayed tumor growth, suggesting a potential target for MDSC inhibition. Clin Cancer Res; 23(9); 2346–55. ©2016 AACR.

The cellular and molecular mechanisms of IL-12-mediated anti-tumor activity have been examined. BALB/c mice bearing established s.c. RENCA or CT26 tumors that were treated daily with IL-12 showed essentially complete tumor regression while tumors in untreated animals grew progressively. Examination of inflammatory gene expression in tumor tissue from treated vs untreated mice revealed the selective expression of IFN-gamma and the IFN-gamma-inducible CXC chemokine IP-10. Immunohistologic analysis demonstrated that tumors from treated mice were heavily infiltrated with CD8+ T cells and Mac-1+ mononuclear cells. Tumor regression in IL-12-treated mice was associated with expression of the lytic effector molecules perforin and granzyme B. These findings support the hypothesis that the anti-tumor function of IL-12 treatment depends upon the induced expression of IFN-gamma by T cells and/or NK cells, the amplification of the immune response mediated by IFN-gamma-induced expression of chemoattractant cytokines, and the IL-12-dependent potentiation of the cytolytic effector function of recruited CD8+ T cells.

Tumors may escape immune recognition and destruction through the induction of apoptosis in activated T lymphocytes. Results from several laboratories suggest that FasL (L/CD95L) expression in tumors may be responsible for this process. In this study of patients with renal cell carcinoma (RCC), we provide evidence for two mechanisms of T-cell apoptosis. One mechanism involves the induction of apoptosis via FasL expression in tumor cells. This is supported by several observations, including the fact that tumor cells in situ as well as cultured cell lines expressed FasL mRNA and protein by a variety of techniques. The FasL in RCC is functional because in coculture experiments, FasL+ tumors induced apoptosis in Fas-sensitive Jurkat T cells and in activated peripheral blood T cells but not in resting peripheral blood T cells. Most importantly, antibody to FasL partially blocked apoptosis of the activated T cells. Moreover, Fas was expressed by T cells derived from the peripheral blood (53% median) and tumor (44.3% median) of RCC patients. Finally, in situ staining for DNA breaks demonstrated apoptosis in a subset of T cells infiltrating renal tumors. These studies also identified a second mechanism of apoptosis in RCC patient peripheral T cells. Whereas these cells did not display DNA breaks when freshly isolated or after culture for 24 h in medium, peripheral blood T cells from RCC patients underwent activation-induced cell death after stimulation with either phorbol 12-myristate 13-acetate/ionomycin or anti-CD3/CD28 antibodies. Apoptosis mediated by exposure to FasL in tumor cells or through T-cell activation may contribute to the failure of RCC patients to develop an effective T-cell-mediated antitumor response.