**3.3 Tumor-associated macrophages**

64 Liver Tumors

cholangiocarcinoma tissue (Chuaysri et al. 2009; Okabe et al. 2009). The origin of these cancer-associated fibroblasts is unknown, although a number of possibilities have been suggested, including hepatic stellate cells (Okabe et al. 2009), portal fibroblasts (Dranoff and Wells 2010) or circulating bone marrow-derived precursor cells (Shimoda et al. 2010). Given the apparent heterogeneous population of cancer-associated fibroblasts observed in cholangiocarcinoma tumors, it is highly likely that these fibroblasts are derived from more than one source. Recently, researchers have performed genetic screening to determine the differences in gene expression between cholangiocarcinoma-derived cancer-associated fibroblasts and non-malignant liver fibroblasts and showed a number of genes associated with angiogenesis, cell proliferation and motility (Utispan et al. 2010). In particular, periostin, a cell adhesion molecule, was shown to be significantly upregulated correlating with shorter survival time in patients and increased cell proliferation and invasive properties *in vitro* (Utispan et al. 2010). Another gene specifically expressed by cholangiocarcinoma-derived cancer-associated fibroblasts is the extracellular matrix protein tenascin-C (Aishima et al. 2003; Iguchi et al. 2009). This gene was expressed predominantly in the stroma near the invasion front of the tumor (Aishima et al. 2003) and was associated with poor prognosis in intrahepatic cholangiocarcinoma (Aishima et al. 2003; Iguchi et al. 2009). Furthermore, the expression of thrombospondin-1 by cancer-associated fibroblasts

correlated with increased metastatsis (Kawahara et al. 1998; Tang et al. 2006).

Fig. 3. Summary of the signalling molecules released by cholangiocarcinoma-derived cancerassociated fibroblasts and their known effects on cholangiocarcinoma progression. CAFs; cancer-associated fibroblasts, HGF; hepatocyte growth factor, SDF-1; stromal derived factor-1. Inflammation and the immune system share a long-standing relationship with tumor initiation and progression. Indeed, the primary risk factor for the development of a number of different tumor types is chronic inflammation of the target organ (Sica 2010). Once a tumor is initiated, tumor-associated macrophages (TAMs) are the major immune cell found within tumors. Macrophages generally have the potential to express and secrete pro- and anti-inflammatory molecules, and as such, may have pro- and anti-tumor activities depending upon the activation stimulus (Sica 2010). For example, macrophages activated with tumor necrosis factor , (considered M1 activation), have anti-tumor activity and signal tissue destruction (Mantovani et al. 2002). Alternatively, in response to interleukin-4, macrophages undergo M2 activation and are involved in tissue repair, remodelling and tumor promotion (Mantovani et al. 2002).

Fig. 4. Schematic representation of the pro- (M1) and anti-(M2) inflammatory activation of macrophages and their effect on tumor growth.

Logically, most TAMs have an M2-like phenotype (Mantovani and Sica 2010) thought to be brought about by various signals expressed within the tumor microenvironment, including interleukin-10, transforming growth factor- and colony stimulating factor-1 (Sica 2010). These signals responsible for the M2-activation of macrophages have been reported to come from myeloid-derived suppressor cells, IL-10+ B lymphocytes, Th2 subtype of T helper cells and the tumor cells themselves (Sica 2010). Once alternatively activated, TAMs exhibit reduced antitumor activities, while increasing the production of mediators of angiogenesis such as VEGF and IL-10 (Mantovani and Sica 2010), as well as M2-specific genes that are known to be involved in promoting cell proliferation (Mantovani and Sica 2010). These events are summarized in Figure 4.

It has been proposed that strategies to inhibit the M2- and activation of the M1-inducing signals may lead to the restoration of the anti-tumor functions of TAM and help to remove the protective signals originating from the M2 TAM (Sica and Bronte 2007), which may trigger an innate immune response, thereby reducing tumor size (Sica 2010).
