IL-6 modulates immune responses and is essential for timely wound healing. As the functions mediated by IL-6 require binding to its specific receptor, IL-6Rα, it was expected that mice lacking IL-6Rα would have the same phenotype as IL-6–deficient mice. However, although IL-6Rα–deficient mice share many of the inflammatory deficits seen in IL-6–deficient mice, they do not display the delay in wound healing. Surprisingly, mice with a combined deficit of IL-6 and IL-6Rα, or IL-6–deficient mice treated with an IL-6Rα–blocking Ab, showed improved wound healing relative to mice with IL-6 deficiency, indicating that the absence of the receptor contributed to the restoration of timely wound healing, rather than promiscuity of IL-6 with an alternate receptor. Wounds in mice lacking IL-6 showed delays in macrophage infiltration, fibrin clearance, and wound contraction that were not seen in mice lacking IL-6Rα alone and were greatly reduced in mice with a combined deficit of IL-6 and IL-6Rα. MAPK activation-loop phosphorylation was elevated in wounds of IL-6Rα–deficient mice, and treatment of wounds in these mice with the MEK inhibitor U0126 resulted in a delay in wound healing suggesting that aberrant ERK activation may contribute to improved healing. These findings underscore a deeper complexity for IL-6Rα function in inflammation than has been recognized previously.

The tumor microenvironment is known to have a profound effect on tumor progression in a highly context-specific manner. We have investigated whether peritoneal inflammation plays a causative role in ovarian tumor metastasis, a poorly understood process. Implantation of human ovarian tumor cells into the ovaries of severe combined immunodeficient mice resulted in peritoneal inflammation that corresponds temporally with tumor cell dissemination from the ovaries. Enhancement of the inflammatory response with thioglycolate accelerated the development of ascites and metastases. Suppression of inflammation with acetyl salicylic acid delayed ascites development and reduced tumor implant formation. A similar prometastatic effect for inflammation was observed when tumor cells were injected directly into the peritoneum of severe combined immunodeficient mice, and in a syngeneic immunocompetent mouse model. Inflammationmodulating treatments did not affect primary tumor development or in vitro tumor cell growth. Depletion of peritoneal macrophages, but not neutrophils or natural killer cells, reduced tumor progression, as assessed by ascites formation and peritoneal metastasis. We conclude that inflammation facilitates ovarian tumor metastasis by a mechanism largely mediated by macrophages, and which may involve stromal vascular endothelial growth factor production. The confirmation of these findings in immunocompetent mice suggests relevance to human disease. Identifying the mechanisms by which macrophages contribute to tumor metastasis may facilitate the development of new therapies specifically targeting immune cell products in the tumor microenvironment. [Cancer Res 2007;67(12):5708-16]

Recent studies have implicated inflammation in the initiation and progression of ovarian cancer, though the mechanisms underlying this effect are still not clear. Toll-like receptors (TLRs) allow immune cells to recognize pathogens and to trigger inflammatory responses. Tumor cell expression of TLRs can promote inflammation and cell survival in the tumor microenvironment. Here we sought to characterize the expression of TLRs in normal human ovaries, benign and malignant ovarian tumors from patients, and in established ovarian tumor cell lines. We report that TLR2, TLR3, TLR4, and TLR5 are strongly expressed on the surface epithelium of normal ovaries. In contrast to previous studies of uterus and endocervix, we found no cyclic variation in TLR expression occurred in murine ovaries. TLR2, TLR3, TLR4, and TLR5 are expressed in benign conditions, epithelial tumors, and in ovarian cancer cell lines. Variable expression of TLR6 and TLR8 was seen in benign and malignant epithelium of some patients, while expression of TLR1, TLR7, and TLR9 was weak. Normal and malignant ovarian stroma were negative for TLR expression. Vascular endothelial cells, macrophages, and occasional fibroblasts in tumors were positive. Functional activity for TLRs was demonstrated by stimulation of cell lines with specific ligands and subsequent activation and translocation of NFkappaB and release of the proinflammatory cytokines interleukin-6 and CCL-2. These studies demonstrate expression of multiple TLRs in the epithelium of normal ovaries and in ovarian tumor cells, and may indicate a mechanism by which epithelial tumors manipulate inflammatory pathways to facilitate tumor progression.

No abstract

No abstract

Supplementary Figure 3 from Macrophages Mediate Inflammation-Enhanced Metastasis of Ovarian Tumors in Mice

Supplementary Figure 1 from Macrophages Mediate Inflammation-Enhanced Metastasis of Ovarian Tumors in Mice

Supplementary Figure 3 from Macrophages Mediate Inflammation-Enhanced Metastasis of Ovarian Tumors in Mice