**6.1 Hepatic stellate cells**

Stromal remodeling occurs routinely during the development of hepatic fibrosis, cirrhosis and HCC, featured with the infiltration of activated hepatic stellate cells (HSC). Upon hepatic injury, HSCs is stimulated and transformed to acquire an activated myofibroblastlike phenotype that is responsible for the excessive hepatic matrix deposition in chronically damaged livers. They are densely located in tumor sinusoids, fibrous septa and HCCgenerated capsule. Activation of HSC is recognized as a key event during hepatotumorigenesis (Zhao et al., 2011).

Activated HSCs considerably increase the activity of NF-κB and ERK in HCC. It is known that both NF-κB and MAP kinase/ERK pathways are involved in the progression of human HCC, and they induce the proliferation of HCC cells, and protect HCC cells from apoptosis (Amann et al., 2009). The paracrinal communication between HSC and HCC forms the major linkage for the induction of HCC development. Several soluble factors secreted by activated HSC are identified to be responsible for the tumorigenic effects. HSC released a substantial amount of protumorigenic factors, including the hepatic growth factor (HGF), which enhances the invasiveness of HCC cells. The growth and the migration capability of HCC were impaired once the binding of HGF to HCC cells was disrupted (Barnaeva et al., 2007). Other studies demonstrated that TGF-β secreted by HSC accelerated tumor progression in neoplastic hepatocyte (Sano et al., 2005). TGF-β was able to induce epithelial to mesenchymal transition and augment PDGF signaling in oncogenic Ras-transformed hepatocyte. It is believed that a combination of HSC-released growth factors consisting of FGF-1 and -2, PDGF and IGF are responsible for promoting HCC tumorigenesis (Bataller & Brenner, 2005). The emerging evidences support that the activated HSC/myofibroblasts in tumor microenvironment have huge impact on HCC development and progression, and this stromal components should be regarded as one of the primary targets in HCC therapy.

#### **6.2 Heparan sulfate proteoglycan modulating enzymes**

Heparan sulfate proteoglycans (HSPG) play important biological roles in both cellular and extracellular context, contributing to the proper communication between cells and their

Novel Therapeutic Targets for Hepatocellular Carcinoma Treatment 45

migration in various HCC cell lines, and enhanced tumor growth in vivo. The tumorigenic effect of SULF2 is partially brought by the induction of the aforementioned pro-cancerous glypican-3 expression. It was found that SULF2 enhanced the binding of FGF2 to the cancer cell and activated FGF signaling in a glypican-3 dependent manner (Lai et al., 2008). In addition, SULF2 increased cell surface glypican-3 and Wnt3a level in HCC, leading to the

SULF2 is a rational target in HCC therapy as suggested by several SULF2 knockdown studies. RNAi-induced suppression of SULF2 reduced the cell growth and migration in cell lines with high SULF2 expression (Lai et al., 2010) in vivo and in vitro. Knockdown of SULF2 was able to reduce the expression of GPC3, as well as the activity of FGF signaling by blocking FGF2 binding (Lai et al., 2008). Reduction of GPC3 also downregulates Wnt3a expression, and attenuates the Wnt/β-catenin signaling with reduced phosphorylated GSK3-β and β-catenin. Given the relationship between SULF2 and GPC3, it is worthwhile to investigate the clinical benefit in targeting SULF2 in HCC treatment. Furthemore, SULF2 protects against caspase 3 and 7 mediated apoptosis induced by PI3K, ERK and JNK inhibitor. Inhibition of SULF2 re-sensitized HCC cells to the drug-induced apoptosis by reducing phosphorylation of AKT, downregulation of cyclin D1 and anti-apoptotic BCL-2, as well as upregulation of pro-apoptotic BAD (Lai et al., 2010). The findings might have

Abnormal epigenetic events are frequently observed in HCC, which can alter gene expression through modification of histone tails or DNA. The major players contributing to these aberrations such as DNA methyltransferase and histone deacetylase are under intensive investigations. In fact, there are many other players involved during the establishment of aberrant epigenetic status. Among them, polycomb repressive complexes (PRC) are catching more attention recently due to their significant roles during cancer development via suppression of various tumor suppressor genes (Steele et al., 2006). In human, there are two polycomb repressive complexes namely PRC1 and PRC2. Despite their unique gene repression mechanism, both of them are frequently involved in the oncogenesis of HCC. Targeting of epigenetic modulators in theory generates persistent effects on tumors as heritable changes are induced. Such an approach is superior to

PRC aroused increasing attention recently as they are shown to contribute heavily in the maintenance of stem cell and the determination of cell fate. BMI1 is a critical component of PRC1 in mediating the ubiquitination of histone in order to regulate local gene expression. BMI1 is not detected in normal hepatocyte but is overexpressed in HCC. Dysregulation of BMI1 is speculated to promote activation of cancer stem cell in HCC. BMI1 has a higher basal level in the side-population (SP) cell where such a subgroup of cancer cells is characterized by the ability to exclude Hoechst 33342 dye via the ABC cassette transporter. This subpopulation is believed to harbor stem cell properties, and BMI1 is shown to play a

increase of glypican-3-dependent Wnt/β-catenin signaling (Lai et al., 2010).

implication to develop combinatory treatment against drug-resistant HCC.

targeting other molecular players that only bring out transient effects.

crucial role in their self-renewal process (Chiba et al., 2008).

**7. Epigenetic modulator** 

**7.1 BMI1** 

surrounding components. While extracellular HSPGs function to maintain extracellular matrix (ECM) self assembly and integrity with other ECM molecules, cell surface HSPGs are responsible for the binding of growth factors, chemokines, cytokines and enzymes. In addition to normal biological process, HSPGs also influence a number of pathological events including inflammation, tumor growth, metastasis and angiogenesis.

### **6.2.1 Heparanase**

Evidences suggested that the expression of heparanase, an enzyme that degrade the side chain heparin sulfate, is closely related to tumor invasion, angiogenesis and metastasis in HCC (El-Assal et al., 2001). Heparanase level is high both in HCC patient serum and tumor tissues. Heparanase level in serum is linked with the aggressiveness of HCC (Wang et al., 2010), and that in tumors is positively correlated with tumor size, staging and portal vein invasiveness (El-Assal et al., 2001). It is speculated that the major pro-tumorigenic effect of heparanase is derived from the ability to cleave HSPG, resulting in the release of HS-bound molecules such as ECM digesting enzymes and angiogenic factors. Consequent ECM degradation and angiogenic factor released combine to construct a microenvironment favorable for HCC cell migration and invasion (Zhang et al., 2007).

Extensive cleavage of heparin sulfate might release other cell surface bound factors such as growth factors and chemokines that potentially generate diverse biological effects in both autocrine and paracine manners. Upregulation of heparanase is associated with increased releasing of basic fibroblast growth factor (bFGF). bFGF released in this way contributes to tumor progression through the activation of oncogenic signaling and construction of a favourable tumor niche (Zhao et al., 2006).

Targeting heparanase provides a novel perspective in managing HCC by modulating the tumor-stromal communication. Knocking down of heparanase can significantly inhibit the invasiveness, metastasis, and angiogenesis of HCC cell both in vitro and in vivo (Zhang et al., 2007). Several molecule inhibitors of heparanase can also attenuate the progression of hepatoma cells. The antitumor effect is possibly generated by preventing the degradation of ECM and basal membrane. Another study showed inhibiting heparanase could effectively stop the release of bFGF so as to inactivate the bFGF signaling effect and suppress subsequent angiogenesis (Zhao et al., 2006). These findings have gradually switched the attention in cancer therapy research, from focusing solely in intracellular targets to the interplay between cancer cells and the surrounding microenvironment.

#### **6.2.2 Sulfatase 2**

Another important feature of heparin sulfate chains is related to its substrate binding capacity. 6-O-sulfation, a type of heparin sulfate modification, is known to play a specific role in modulating ligand binding. The enzyme SULF2 is a member of the sulfatase family that modulates critical cellular signaling pathways by the removal of 6-O-sulfation (Morimoto-Tomita et al., 2002). In contrast to another sulfatase member tumor suppressor SULF1, SULF2 has an oncogenic role in cancer, and its expression is elevated in HCC. Upregulation of SULF2 is observed in 57% HCC tissues and 73% HCC cell lines. Level of SULF2 is positively correlated with a more aggressive tumor phenotype and poorer patient survival (Lai et al., 2008). Ectopic expression of SULF2 promoted cell proliferation and migration in various HCC cell lines, and enhanced tumor growth in vivo. The tumorigenic effect of SULF2 is partially brought by the induction of the aforementioned pro-cancerous glypican-3 expression. It was found that SULF2 enhanced the binding of FGF2 to the cancer cell and activated FGF signaling in a glypican-3 dependent manner (Lai et al., 2008). In addition, SULF2 increased cell surface glypican-3 and Wnt3a level in HCC, leading to the increase of glypican-3-dependent Wnt/β-catenin signaling (Lai et al., 2010).

SULF2 is a rational target in HCC therapy as suggested by several SULF2 knockdown studies. RNAi-induced suppression of SULF2 reduced the cell growth and migration in cell lines with high SULF2 expression (Lai et al., 2010) in vivo and in vitro. Knockdown of SULF2 was able to reduce the expression of GPC3, as well as the activity of FGF signaling by blocking FGF2 binding (Lai et al., 2008). Reduction of GPC3 also downregulates Wnt3a expression, and attenuates the Wnt/β-catenin signaling with reduced phosphorylated GSK3-β and β-catenin. Given the relationship between SULF2 and GPC3, it is worthwhile to investigate the clinical benefit in targeting SULF2 in HCC treatment. Furthemore, SULF2 protects against caspase 3 and 7 mediated apoptosis induced by PI3K, ERK and JNK inhibitor. Inhibition of SULF2 re-sensitized HCC cells to the drug-induced apoptosis by reducing phosphorylation of AKT, downregulation of cyclin D1 and anti-apoptotic BCL-2, as well as upregulation of pro-apoptotic BAD (Lai et al., 2010). The findings might have implication to develop combinatory treatment against drug-resistant HCC.
