The glypicans, a family of proteins classified as HSPGs, have been shown to interact with a number of growth factors and modulate growth factor activity (1) and are linked to the extracellular side of cell membrane by a glycosylphosphatidylinositol (GPI) anchor (2). Members of this large family of transmembrane proteins have been identified in both mammals and drosophila: 6 glypicans (GPC-1 through GPC-6) in mammals, and two others in the fly (2). Glypicans have a core protein size of ~60–70 kDa and express an N-terminal secretory signal peptide along with a hydrophobic domain that is used for the addition of the GPI anchor at the C-terminus. Another characteristic that is shared by all glypicans is the location of the insertion sites for the heparan sulfate chains (HSC), which seems to be restricted to the last 50 amino acids in the C-terminus, placing the chains close to the cell membrane. Additionally, the position of 14 cysteine residues is conserved, further strengthening the structural relationship of the proteins within the family (2).

Much of the literature on GPC-3 stems from its proposed role in vivo as well as in vitro in cancer models. Multiple studies have shown a correlation between GPC-3 and hepatocellular carcinoma (HCC) due to the nature of the protein being expressed during normal cellular growth and movement in hepatocytes (1,3–6). In patients who received surgical treatment for HCC, a decrease in GPC-3 protein was observed (3). In addition, in cells derived from hepatocellular carcinoma GPC-3 has been observed to be secreted into culture media (7). GPC-3 does not seems to be expressed by other liver cell types (hepatic stellate cells, kupffer cells, bile duct cells, endothelial cells and fibroblasts) but it appears to be expressed exclusively by hepatocytes. It is well known only that its expression from hepatocytes influences the activity and function of non-parenchymal liver cells.

GPC-3 was initially hypothesized to interact with insulin-like growth factor in selected cell types, though this notion has been challenged as this may not be the case in all cell types (8). In a study done on the loss-of-function mutation of GPC-3 in embryonic development, researchers observed an overgrowth of organs in fetal and postnatal development, known as Simpson-Golabi-Behmel overgrowth syndrome (9). This evidence cast GPC-3 into a strict role as a tumor suppressor. In another study looking at the metastatic adenocarcinoma mammary LM3 cell line, GPC-3 was shown to inhibit the canonical Wnt signals involved in cell proliferation and survival. In the same study, GPC-3 was also observed to play a role in activating the non-canonical Wnt pathway, which directs cell morphology and migration (8). Furthermore, GPC-3 was also found to act as a potential tumor suppressor in lung tissue (10). This has re-opened the case of GPC-3 as a strict suppressor of cell growth and introduced the possibility of its role in stimulating growth and even tumorigenesis in a tissue-dependent manner in some cell lineages.

This being said, certain key elements have been observed which give reason for upregulation of GPC-3 in certain cell types and not others. GPC-3 is able to bind Wnt and Hedgehog (Hh) signaling proteins and was also shown to have the ability to bind basic growth factors such as fibroblast growth factor 2 (FGF2) through its heparan-sulfate glycan chains (11). Additionally, expression of sulfatase 2 (SULF2), another tumor marker expressed in hepatocytes, was found to be correlated with increased expression of GPC-3. When GPC-3 was decreased, FGF2 binding was decreased in SULF2 expressing hepatocellular carcinoma cells. SULF2 expression in resected HCC tissue was also shown to have a worse prognosis and a higher rate of recurrence after surgery (12).

Herein we synthesize the literature on GPC-3, including its structure, expression in normal tissues, as well its role as a serum marker as well as possible therapeutic target. The review also includes current knowledge on GPC-3 expression as a prognostic marker in HCC and its role in liver regeneration post partial hepatectomy (PH). In addition, we discuss embryonic and adult tissue expression of GPC-3. Lastly, we report the interactions of GPC-3 with CD81 as CD81 has been shown to interact with GPC-3 as a binding partner (13,14).

GPC-3 may be a potential serum marker in diseases such as HCC in which expression of GPC-3 is markedly increased (20,21). Capurro et al examined GPC-3 protein expression and serum levels using immunohistochemistry and ELISA in HCC patients, as well as serum levels in healthy donors and patients with hepatitis and liver cirrhosis. Results revealed increased GPC-3 expression levels in patients with HCC, but not in healthy hepatocytes. GPC-3 serum levels were significantly elevated in HCC patients, but undetectable in the serum of healthy and hepatitis-infected patients (22).

Discussing where this protein is cleaved, the cleaved portion of the NH(2)-terminal was found to be between Arg(358) and Ser(359) of GPC-3. In addition to this, it was observed that soluble GPC-3 can be specifically detected in the sera of patients with HCC using N-mAbs (23). Unfortunately, the cleavage events and other post-translational modifications are understudied. It is necessary to investigate in terms of function and cellular localization the cleavage events on GPC-3. Focusing on serum marker it is necessary to understand if the protein is cleaved and which portion is secreted improving the knowledge of secretory pathways involved in its sorting in and out of the cells, with major attention on exosomal cargos.

GPC-3 was also identified as a novel diagnostic marker for human melanoma it its early stages. In patients with melanoma, GPC was detected. However, the protein was not found in the sera of healthy patients, or those with benign skin lesions (large congenital melanocytic nevus). It was also detected in the serum of patients with stage 0, in situ melanoma (28).

In another study carried out to observe biomarkers for HCC, preoperative serum GPC-3-N (sGPC-3N) levels were measured alongside with serum AFP in patients with HCC. It was observed that high serum AFP corresponded with a high sGPC-3N level. The study states that sGPC-3N may serve as an independent prognostic biomarker in HCC patients (24).

Serum α-fetoprotein is the common diagnostic marker to detect HCC. Several studies showed that the combination of α-fetoprotein and GPC-3 increase significantly the sensitivity of diagnostic value compared to their sensitivity analyzed independently (43). Several markers are under investigation to ameliorate the sensitivity and specificity of HCC detection such as: des-γ carboxyprothrombin, lens culinaris agglutinin-reactive α-fetoprotein, Golgi protein 73 (43,44). However, GPC-3 seems to be more specific due to its peculiar expression in hepatocytes as showed before, and this represents the main advantage for its use in HCC detection even if its mechanism of transport in the serum is still unknown. For its peculiar localization and post-translational modifications GPC-3 could represent an ideal target for antibody therapeutic approach with the possibility to bind its membrane form thus inhibiting its activity. Moreover, a chemical approach could obtain the cleavage events that determine GPC-3 functional modulation and cellular localization. Consequent to its expression in pre-neoplastic hepatocytes, this offers the unique advantage of allowing an early and specific therapeutic approach.

Few studies have demonstrated the expression of GPC-3 pre- and post-hepatic resection. Due to hepatic progenitor cells expressing cytokeratin 19 (CK19) and GPC-3 at varying phenotypes, cell samples were analyzed from patients who had a liver resection and were statistically compared with each phenotype of expression in Table III. It was found that CK19⁺/GPC-3⁺ HCC was the most aggressive subtype, followed by the CK19⁻/GPC-3⁺ HCC and finally with the CK19⁻/GPC-3⁻ HCC subtype being the least aggressive from the subtypes observed (45). Therefore, a poorer prognosis may be associated with patients undergoing a hepatectomy who have the expression of both CK19 and GPC-3.

Liu et al, on the other hand, described transgenic GPC-3 mice, which were under the control of the albumin promoter gene, overexpressed GPC-3 and actually had a suppression of hepatocyte proliferation and liver regeneration following partial hepatectomy (30,46). These mice developed normally and the authors claim that GPC-3 may play a negative regulatory role in hepatocyte proliferation, showing a contrasting role of GPC-3 in the hepatocytes. Further investigation using models post-PH is needed to uncover various important aspects of GPC-3 expression such as the density and location of GPC-3 expression (i.e. the cut margin or other location). This research may be instrumental in further elucidation of the role and mechanisms of GPC-3 in liver growth and regeneration post-resection and even injury (mechanically, inflammatory, or otherwise). However, limited studies have been conducted on the regeneration of the liver.

At this time there is no evidence on a possible function and presence of GPC-3 in other liver diseases such as fibrosis, fatty liver disease or liver cirrhosis. The peculiarity of its expression in neoplastic hepatocytes makes GPC-3 not only a unique HCC marker but also a possible therapeutic target. However, studies in cirrhotic livers without HCC are necessary to rule out the role of GPC-3 or the presence in these circumstances.

Therapeutic targets are especially important in the treatment of various types of cancer. GC33, which is a humanized monoclonal antibody, was shown to bind human GPC-3. GC33 has been observed to contain antitumor properties and targets GPC-3 specifically. It was noted that GC33 was well tolerated in advanced HCC and provided some benefit in the clinical treatment (47). In addition, human (MDX-1414 and HN3) and humanized mouse (GC33 and YP7) antibodies that also target GPC-3 are under different stages of clinical development and could also aid in the treatment of HCC (21). miR-219-5p is a microRNA (miRNA) which could exert tumor suppression on hepatic cells expressing GPC-3. miR-219-5p reduced both the mRNA and protein levels of GPC-3 and exerted tumor-suppressive effects in HCC (48). Finally, nanoparticles are also under development to target and bind GPC-3 in cells that could provide useful for further imaging and targeting of GPC-3 (49).

Research has also been conducted in regards to peptide sequences which bind GPC-3. Feng et al described a 12-mer peptide sequence (DHLASLWWGTEL) which was able to act as a probe and bind GPC-3 for HCC detection (21). The peptide allows visualization of the specific sequence via near-infrared fluorescence (50). Another therapeutic target on GPC-3 is mir717, a miRNA, which is located on intron 3 of GPC-3 on the X chromosome. It has been shown to play a regulatory role in renal osmoregulation. mir717 might be connected in some way with obesity regulation, one of the risk factors of HCC (51,52).

It has been proposed that GPC-3 acts as a negative regulator in the Hedgehog signaling pathway during development and as a result, a non-functional GPC-3 protein could be the cause of overgrowth or a GPC3 modified by post-translational modifications. GPC-3 has also been associated with the binding of Wnt to its signaling receptor, Frizzled (53–56). Both the Hh and canonical/non-canonical pathways of Wnt could be potential therapeutic targets as GPC-3 may play a role in proliferation and/or growth suppression, depending on the type of tissue and the stage of development (8,57). In addition, the peptide sequences Arg-Leu-Asn-Val-Gly-Gly-Thr-Tyr-Phe-Leu-Thr-Thr-Arg-Gln and Tyr-Phe-Leu,Thr-Thr-Arg-Gln showed selective binding of GPC-3 (58) which could provide useful for a target that is needing to bind onto GPC-3 for regulation.

GPC-3 has been observed to interact with a number of co-receptors that further modulate cellular expression. Liu et al showed that GPC-3 and CD81 levels were significantly upregulated in general in the transgenic mice following partial hepatectomy (30,46). They showed that the negative regulatory role was somehow associated with GPC-3 and CD81 and that there is enhanced association between GPC-3 and CD81. GPC-3 and CD81 have both been associated as binding partners in which expression could influence the Hh (13). Co-localization of GPC-3 and CD81 was also shown to occur 2–6 days after hepatectomy (30). In contrast, another study observed that GPC-3 binds to members of the Hh pathway and prevents their interactions with the patched-1 receptor. There was decreased binding of GPC-3 with Hh and CD81 following PH and GPC-3/CD81 may play a role in the termination of liver growth following liver regeneration (14,46). Dysregulation of the association between GPC-3/CD81, which may occur during hepatitis C infection can result in dysregulated signaling and proliferation in infected hepatocytes (59). The structure and character of GPC-3 as a transmembrane protein, in conjunction with the current literature, lends itself to the strong possibility of many associations or interactions with other co-receptors and proteins, not only in hepatocytes, but possible in other normal or cancerous tissue types. It is important to emphasize again the paucity in the current literature in regard to this, and many other aspects of GPC-3.

GPC-3 is a molecule that is still not fully understood in its role in the proliferation and suppression of cell growth in normal and abnormal or cancerous tissue and also in structural and post-translational modifications. The question still remains as to why it is normally expressed in some tissues, while remaining silenced in others. Future research that looks into the upstream and downstream cell signaling pathways and how GPC-3 may be involved could provide further answers. In addition, studies investigating a more complete picture and analysis of GPC-3 structure are warranted, as glycosylation, sulfonation, or other structural components of the protein may be important in understanding its regulatory function within differing tissue types. In this same vein, continued insight into the cleavage of GPC-3 and the function of the cleaved versus non-cleaved form of the molecule could lead to a more complete comprehension of function. Other research paths may include developing the theory of GPC-3 as a dual tumor suppressor and oncogene, as dependent upon its structure. With exosome research brightening the horizons, more attention and detail must be paid to uncover the importance of exosomal GPC-3 as a serum marker or as possible therapeutic target in HCC. GPC-3 is a molecule that could further connect the missing links in liver cancer research and lead to an abundance of new intercellular relationships to reveal important aspects of the biology of this disease.