**3. Cell-surface protein**

#### **3.1 Glypican-3**

Glypican-3 (GPC3) is a protein anchored to the cell surface by a glycosylphosphatidylinositol link. Glypican-3 is highly expressed in HCC, and plays a role in stimulating various tumorigenic signaling pathways. GPC3 is specifically expressed in HCC, but not in cholangiocarcinoma or normal liver tissue. More than 70% of HCC tumors were observed with high GPC3 level compared to normal liver tissues (Hsu et al., 1997). Consistent with the high GPC3 protein expression found in clinical samples, numerous

Novel Therapeutic Targets for Hepatocellular Carcinoma Treatment 39

strongly correlated with the prognosis of HCC patients, predicting a shorter overall survival rate as well as higher relapse rate and venous infiltration after surgery (Wang et al., 2005). This CDH17 isoform, together with others that are exclusively expressed in HCC, might contribute to the pathogenesis of HCC. Strategy to target the isoforms of CDH17 allows specifically aiming malignant cells rather than the normal hepatocytes. All available evidences suggested

Arginine content is well-known to affect transplanted tumor in mice. Enhanced in vivo tumor growth is observed when mice were fed with diet rich in arginine. On the other hand, depletion of arginine from their diet inhibits the growth of metastatic tumor (Gonzalez & Byus, 1991). It is later proved that arginine is essential for the survival of cancer cells. Cancer cells are dependent on exogenous arginine for growth because most of them cannot synthesize their own and become auxotropic for arginine (Dillon et al., 2004). There are various explanations for the acquisition of arginine auxotropic phenotype in various cancer cells, but generally it is associated with the downregulation of argininosuccinate synthase (ASS) (Dillon et al., 2004). Arginine auxotrophy is also a common phenomenon in HCC cells

In somatic cells, deficiency of arginine puts cell cycle on hold, and cells enter the quiescence G0 phase. They can tolerate the depletion of arginine for weeks and return to normal condition once the arginine content is resumed. On the other hand, arginine deficiency is not sustainable in cancer cells (Delage et al., 2010). Defect in cell cycle checkpoint drives continuous cell proliferation even with insufficient arginine, but arginine is necessary for metabolic and enzymatic pathways in malignant cells. In essence, cancer cells with shortage of ASS rely heavily on exogenous arginine. If the uptake of arginine is disrupted, or the stability of arginine is lowered, cell death will occur due to a loss of gross balance (Delage et al., 2010). This physiological difference between normal and cancer cell makes the arginine metabolic pathway a potent target in treatment to distinguish HCC cells from normal cells. Reducing arginine stability is one of the strategies against malignant cells, and argininedegrading enzyme is the major group of enzymes that can serve the purpose in depleting internal arginine. Arginase belongs to such group of enzyme which is responsible for arginine degradation in the urea cycle, and its anticancer effect is well documented (Bach et al., 1963). In addition to arginase, the enzyme arginine deiminas is proved to efficiently deplete cellular arginine in vitro and in vivo (Cheng et al., 2007). Recombinant argininedegrading enzymes were developed, and their anticancer effect was investigated in HCC. Satisfactory result was obtained using recombinant arginase and arginine deiminas to combat ASS-deficient tumors (Izzo et al., 2004). Studies are conducted to improve the efficacy of these arginine-degrading enzymes. Modification such as pygelation can increase the half-life of the enzyme and prolong its activity. Phase III trial deploying a pegylated form of recombinant ADI is undertaken in HCC patients who have failed prior systemic treatment. It is also reported that a modified recombinant human arginase is able to inhibit

ASS-positive HCC, and inhibit tumor cell growth (Cheng et al., 2007).

that targeting CDH17 might be a prospective molecular-based therapy in HCC.

**4. Metabolic pathway 4.1 Arginine metabolism** 

due to their lackage of ASS (Ensor et al., 2002).

HCC cell lines have high expression level of GPC3 (Midorikawa et al., 2003). In addition, GPC3 expression is correlated with the prognosis of HCC, where GPC3-positive HCC patients have a significantly lower 5-year survival rate than patients who are GPC3-negative (Shirakawa et al., 2009).

One of the GPC3 tumorigenic roles is the activation of Wnt/β-catenin signaling. It is shown that GPC3 is able to interact with Wnt ligands, and induces canonical Wnt-signaling to trigger the stabilization of β-catenin and induction of cyclin D1 (Capurro et al., 2005). The heparin sulfate chain of GPC3 is reported to bind with basic growth factors such as FGF-2. The interaction between GPC3 and FGF-2 is frequently observed in HCC cells, and is responsible for phosphorylation of ERK and AKT (Midorikawa et al., 2003). This interaction plays a role in the increase of HCC cell proliferation, and growth of tumor in nude mouse model. Additionally, GPC3 interplays with hedgehog signaling in regulating developmental growth (Capurro et al., 2008). Though yet to be elucidated, the GPC3-hedgehog signaling is suggested to contribute to HCC development.

Targeting GPC3 and its related growth signaling is a relevant approach to inhibit HCC growth. Inhibition of the interaction between GPC3 and Wnt or FGF-2 should theoretically reduce HCC growth (Capurro et al., 2005; Midorikawa et al., 2003). GPC3 is also a useful target in immunotherapy against HCC. The therapeutic monoclonal antibody against GPC3 has been developed which could induce antibody-dependent HCC cytotoxicity. Targeting GPC3 is able to inhibit tumor growth of HCC cell line xenograft (Ishiguro et al., 2008). Study also showed the concomitant treatment with GPC3 monoclonal antibody and sorafenib was more potent in preventing tumor growth than sorafenib alone in the HepG2 xenograft model (Ishiguro et al., 2008). It is likely that targeting GPC3 could provide great clinical benefit during HCC management.

#### **3.2 Cadherin 17**

Cadherins are important cell adhesion molecules strongly associated with cancer progression. Downregulation of E-cadherin (Du et al., 2009) and overexpression of Pcadherin are often observed in advanced tumor which processes crucial cellular event like epithelial-mesenchymal transition (Sun et al., 2011). Cadherin 17 (CDH17) is another adhesion molecule upregulated in HCC, and it is linked to the tumorigenesis in various gastrointestinal regions (Wang et al., 2005). The upregulation of CDH17 is capable of transforming premalignant liver progenitor cells into liver carcinomas in mice. While forced expression of CDH17 promoted tumor growth from hepatic progenitor cells, silencing of CDH17 reduced the aggressiveness of metastatic HCC cells (Liu et al., 2009). Knockdown of CDH17 by RNA-interference decreased the proliferation rate of HCC cell lines despite their metastatic potential in vitro and in vivo. It is shown that targeting CDH17 can concurrently inactivate Wnt/β-catenin signaling and reduce cyclin D1 level, leading to both growth inhibition and cell death. Inhibition of CDH17 results in the re-localization of nuclear βcatenin to the cytoplasm so as to attenuate the Wnt/β-catentin signaling (Liu et al., 2009).

Multiple isoforms of CDH17 protein are present in the HCC samples, and it is found that the isoform lacking exon 7 is the most abundant in HCC samples (Wang et al., 2005). CDH17 isoform lacking exon 7 cannot be found in normal liver tissue whereas it is present in about 50% of human HCC and 30% of premalignant tissues. Detection of this CDH17 isoform was strongly correlated with the prognosis of HCC patients, predicting a shorter overall survival rate as well as higher relapse rate and venous infiltration after surgery (Wang et al., 2005). This CDH17 isoform, together with others that are exclusively expressed in HCC, might contribute to the pathogenesis of HCC. Strategy to target the isoforms of CDH17 allows specifically aiming malignant cells rather than the normal hepatocytes. All available evidences suggested that targeting CDH17 might be a prospective molecular-based therapy in HCC.
