Hepatocellular carcinoma (HCC) represents the second most common cause of death from
\n cancer worldwide, and was responsible for nearly 746 000 deaths in 2012 1]\u00e2\u20ac\u201c3]. In patients with cirrhosis, HCC is the most common cause of death. Worldwide, chronic
\n hepatitis B virus infection remains the major risk factor, with 80\u00c2\u00a0% of cases occurring
\n in eastern Asia and sub-Saharan Africa. In most countries, the mortality rate of HCC
\n approximates the incidence, which is increasing 4]\u00e2\u20ac\u201c6]. This is partly due to the rising prevalence of advanced fatty liver disease and
\n chronic hepatitis C, alongside other risk factors such as hepatitis B infection and
\n alcohol-related cirrhosis. Some progress has been made with prevention, for example
\n emerging antiviral agents and vaccination for hepatitis B. However, the vast majority
\n of HCC cases are associated with fibrosis, and 90\u00c2\u00a0% of tumours develop in cirrhotic
\n livers 4], 5], 7]\u00e2\u20ac\u201c10]. Furthermore, liver disease severity markers correlate with tumour formation 4]\u00e2\u20ac\u201c6], 9], 11]\u00e2\u20ac\u201c14]. Currently there are no effective anti-fibrotic therapies available to halt the fibrosis-cirrhosis-HCC
\n continuum. Patients who present with early disease may benefit from resection, transplantation
\n or loco-regional therapy, however many are unsuitable for curative treatment due to
\n advanced malignancy, or the severity of co-existing liver disease. The multi-tyrosine
\n kinase inhibitor sorafenib is the only available systemic chemotherapy agent with
\n survival benefit for advanced stage HCC, however its use is limited to those with
\n well-preserved liver function 11]. Whilst there is scope to optimize our use of existing treatments, for example by
\n targeting tumours earlier and combining local and systemic approaches, efforts to
\n broaden our chemotherapy armamentarium have been disappointing. Numerous molecular
\n therapies with robust preclinical evidence for efficacy have failed to show benefit
\n in clinical trials. This may in part reflect the abnormal tumour microenvironment,
\n which acts to support the persistence and growth of cancer cells, and has resulted
\n in the peri-tumoural stroma and its cellular inhabitants becoming an intense area
\n of study in the search for efficacious therapies for HCC.\n <\/p>\n
In this review we focus on the complex interplay between hepatic stellate cell (HSC)
\n biology and hepatocarcinogenesis. The mechanisms by which HSC may facilitate HCC development
\n and progression are likely to involve diverse biological processes including regulation
\n of extracellular matrix (ECM) turnover, growth factor and cytokine signalling, promotion
\n of tumour angiogenesis and immunomodulation. We will discuss how this burgeoning area
\n of research may yield exciting new therapies for patients with HCC.\n <\/p>\n
The stroma is a central component of both hepatic fibrosis and carcinogenesis, and
\n is a key player in the cellular and molecular mechanisms linking these processes.
\n It is still unclear, however, whether liver fibrosis specifically promotes HCC, or
\n if it is merely a wound-healing by-product of chronic hepatic injury and inflammation,
\n with no direct impact on liver cancer formation 8], 13]\u00e2\u20ac\u201c15]. Evidence would suggest the former; the identification of gene signatures from non-tumoural
\n tissue correlating with late recurrence of HCC, supports the concept of a \u00e2\u20ac\u02dcfield effect\u00e2\u20ac\u2122
\n in cancer development 9], 11], 13], 14], 16]\u00e2\u20ac\u201c25].\n <\/p>\n
Following liver injury, quiescent HSC become activated to matrix-secreting myofibroblasts It is well-known that activated HSC infiltrate HCC stroma and peri-tumoural tissue, Consisting of an ?- and ?-subunit, integrins form a family of transmembrane receptors In activated HSC downstream integrin signalling, via the focal adhesion kinase (FAK)-phosphatidylinositol The ?1 integrin subfamily has been extensively studied in the context of HCC, and Other integrin subunits, in addition to ?1, have been reported to have key roles in Crosstalk between integrins and TGF-? signalling has also been studied in hepatocarcinogenesis. HSC have been shown to favour HCC tumourigenicity, potentially as a result of a change HGF is expressed by HSC and myofibroblasts, 77], 78] and is a highly potent hepatocyte growth factor regulating cell proliferation, migration, The pro-tumourigenic activity of fibroblast-secreted HGF has also been reported in vitro.<\/em> Conditioned media from isolated and activated HSC, pre-incubated with anti-HGF antibodies, However, HGF signalling is not unidirectional. A high level of bi-directional crosstalk The large latent TGF-? complex is secreted by most cell types, including human HSC The survival and malignancy of HCC cell lines, including Huh7 and HepG2, have been Therefore, as the role of TGF-? in HCC pathogenesis appears to be highly context-dependent, The gut microbiome is increasingly recognized as a powerful modulator of fibrosis, Angiogenesis has a critical role in HCC initiation, progression and metastasis, as HSC are known to secrete VEGF as well as other angiogenic factors including PDGF, Conditioned media from HCC cells can activate HSC and stimulate VEGF production. Coulouarn Lin et al.<\/em> have also shown increased angiogenesis by activated HSC in vitro<\/em> using a murine HCC cell line (H22) and rat colon microvascular endothelial cells Of particular interest in HCC is the interaction between malignant hepatocytes, endothelial Recently, it has been shown that metformin inhibits angiogenesis in vitro<\/em>, in an HCC (HepG2 line) and HSC (LX2) co-culture system 131]. This was associated with reduced VEGF production. It was postulated that metformin Some of the factors mediating crosstalk between HSC and HCC are summarised in Fig.\u00c2\u00a01.<\/p>\n Fig. 1.<\/strong> Crosstalk between HSC and HCC. HSC-secreted factors such as HGF may promote hepatocarcinogenesis. Tumour immune evasion is now regarded as a hallmark of cancer progression and is therefore The immunosuppressive activities of HSC have only recently been recognised with studies Co-transplantation of HCC cells and HSC into immunocompetent mice promoted HCC proliferation Investigation into the mechanisms underlying HSC immunomodulatory effects in HCC has Three monoclonal antibodies against PD-1, and one against B7-H1 have been developed Fig. 2.<\/strong> Immunomodulatory effects of HSC in HCC. Intratumoural HSC (iHSC) promote HCC progression HSC represent a small percentage of cells within the liver, and specific therapeutic If a targetable, HSC-dependent pathway driving hepatocarcinogenesis is identified, To overcome these problems, a number of groups have explored active targeting of HSC Carriers recently employed have included an antibody to the synaptophysin receptor With these challenges in mind, Bansal et al.<\/em> subsequently developed a recombinant protein construct to deliver interferon gamma Fig. 3.<\/strong> Therapeutic approaches to targeting HSC. HSC have been targeted by coupling a compound Hepatocellular carcinoma (HCC) represents the second most common cause of death from cancer worldwide, and was responsible for nearly 746 000 deaths in 2012 1]\u00e2\u20ac\u201c3]. In patients with cirrhosis, HCC is the most common cause of death. Worldwide, chronic hepatitis B virus infection remains the major risk factor, with 80\u00c2\u00a0% of cases occurring in eastern […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[],"tags":[],"class_list":["post-13607","post","type-post","status-publish","format-standard","hentry"],"_links":{"self":[{"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/posts\/13607","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/comments?post=13607"}],"version-history":[{"count":0,"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/posts\/13607\/revisions"}],"wp:attachment":[{"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/media?parent=13607"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/categories?post=13607"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/healthmedicinet.com\/news\/wp-json\/wp\/v2\/tags?post=13607"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n and are the major source of ECM proteins during liver fibrogenesis 8], 13], 26]. As master regulators of the fibrotic matrix, HSC may therefore directly influence
\n HCC formation via effects on the tumour stroma. Furthermore, it is well established
\n in other systems that complex intercellular signalling networks exist between tumours
\n and cancer-associated fibroblasts, contributing to cancer initiation, growth and progression
\n 8], 13], 16]\u00e2\u20ac\u201c19], 21]\u00e2\u20ac\u201c26]. Tumour secretion of cytokines such as transforming growth factor-? (TGF-?), stimulate
\n myofibroblast activation leading to profound changes in ECM composition and organization.
\n Therefore, HSC or HSC-secreted products may be either permissive or necessary for
\n oncogenesis and HCC persistence. In other cancers, the identification of pathways
\n that the tumour depends upon for growth and proliferation, so-called \u00e2\u20ac\u0153oncogenic addiction
\n loops\u00e2\u20ac\u009d has led to revolutionary therapeutic approaches. The landmark discovery of
\n the protein kinase oncogene BCR-ABL and subsequent development of imatinib, allowed
\n curative treatment of chronic myeloid leukaemia, and has paved the way for targeted
\n therapies in other malignancies 27], 28]. Despite extensive genomic profiling of HCC, targeting other non-kinase oncogenes
\n such as RAS and MYC has proven more challenging. The identification of promising candidate
\n pathways targeting inhibition of a driving molecular alteration, which is also applicable
\n in a significant proportion of patients, remains an elusive yet alluring goal 29]. Furthermore, the microenvironment may modulate susceptibility to inhibition of specific
\n oncogenic pathways. Straussman et al.<\/em> developed a co-culture system to test the ability of 23 stromal cell types to influence
\n the susceptibility of 45 different cancer cell lines to 35 therapeutic agents 7]. They demonstrated that stroma-mediated resistance to anti-cancer drugs (especially
\n targeted agents) is common. In particular, although melanomas expressing mutant BRAF
\n respond to vemurafenib, hepatocyte growth factor (HGF) secretion by peri-tumoural
\n stromal cells correlated with resistance to vemurafenib-induced cell death 7], 30], 31]. This illustrates the importance of stroma-derived resistance to chemotherapy, in
\n many different organs and disease settings. Therefore, in the search for key driver
\n mutations in HCC, the effect of the microenvironment cannot be underestimated. This
\n may necessitate combinations of chemotherapeutic agents, to neutralize specific stromal
\n interactions, resulting in greater overall clinical efficacy.\n <\/p>\nHSC in HCC<\/h4>\n
\n and are localised around tumour sinusoids, fibrous septae and the tumour capsule 32]\u00e2\u20ac\u201c34]. Activated HSC have also been identified around the periphery of dysplastic nodules
\n within the liver 35]. Following activation to the myofibroblast phenotype, HSC secrete substantial amounts
\n of ECM proteins into the stroma. Fibrotic matrix deposition and degradation by HSC
\n is tightly regulated in the liver. For example, tissue inhibitors of metalloproteinases
\n 1 (TIMP-1) secretion favours scar deposition by inhibiting the endogenous matrix-degrading
\n activities of various matrix metalloproteinases (MMPs). However, the balance of TIMPs
\n and MMPs is complex; activated HSC are also a major source of MMP-2 in vitro<\/em>, elevation of which has been correlated with increased tumoural collagen I, extracellular
\n remodeling, and HCC progression 12], 36], 37]. Interestingly, the biomechanics of the ECM are also relevant. Differentiation of
\n primary hepatocytes is inhibited by culture on a stiff collagen gel, with accompanying
\n promotion of proliferation 38], 39]. In vitro<\/em> increasing matrix stiffness has also been shown to directly stimulate growth of the
\n HCC cell lines, HuH-7 and HepG2, and reduce chemotherapy-induced apoptosis 40]. Integrin ?1 signalling was an integral driver of this response, via Fak, Erk, Pkb\/Akt
\n and Stat3 pathways 40]. Furthermore, stromal stiffness is self-perpetuating, causing stellate cell activation,
\n and therefore further fibrosis 15], 41], 42]. Data in humans support these experimental findings. Ultrasound elastography has
\n demonstrated that measurements of liver stiffness predict HCC development 43]\u00e2\u20ac\u201c46]. Similarly, established HCC demonstrates further increases in matrix stiffness, more
\n so than the peri-tumoural hepatic parenchyma 47]. The mechanical tension provided by an altered ECM is likely to act on HCC development
\n and progression via outside-in signalling, for example by integrins, (discussed below)
\n to support tumour growth and progression. This has also been observed in other malignancies,
\n such as a mouse model of breast cancer 48]. Hepatocarcinogenesis in the context of cirrhosis, however, is a unique model of
\n diseased ECM, and an ideal setting to further characterise and potentially target
\n stromal drivers.\n <\/p>\nIntegrins as mediators of HSC\/HCC crosstalk<\/h4>\n
\n that \u00e2\u20ac\u02dcintegrate\u00e2\u20ac\u2122 the extracellular and intracellular environments through binding
\n ECM and the cytoskeleton 49]. Via transduction of signals between the internal and external cellular domains,
\n integrins regulate cell adhesion, spreading, migration, proliferation and differentiation
\n as well as ECM deposition and remodelling 50].\n <\/p>\n
\n 3-kinase (PI3K)-Akt signaling pathway, promotes ECM deposition 51]. Increased ECM stiffness in vitro<\/em> enhances integrin expression and activity and focal adhesion formation, 48] with subsequent activation of downstream integrin signalling within the hepatocyte
\n that may nurture the growth and survival of precancerous cells. Matrix stiffness has
\n been reported to dictate differentiation and chemotherapeutic resistance of human
\n HCC cell lines, with softer matrices abrogating hepatoma proliferation and stiffer
\n platforms promoting proliferation 40], 52]. In an elegant in vivo<\/em> study, cells from the HCC cell line McA-RH7777 were implanted into rats treated with
\n carbon tetrachloride (CCl4<\/sub>) for varying lengths of time, thereby modelling tumourigenesis on different liver
\n stiffness backgrounds. Microarray analysis of the tumours demonstrated a positive
\n correlation between matrix rigidity and tumour angiogenesis 52]. Correlations between collagen expression, integrin expression and tumourigenicity
\n have also been reported in human HCC and murine HCC models 53], 54]. Characterisation of integrin expression in hepatoma cell lines has revealed a high
\n degree of heterogeneity in integrin expression 55]. Comparing two clinically relevant mouse models of HCC, platelet-derived growth factor
\n (PDGF)-C overexpressing and PTEN null mice, Lai et al.<\/em> demonstrated that each model had a specific pattern of integrin gene expression,
\n further indicating HCC heterogeneity 54].\n <\/p>\n
\n hepatocarcinogenesis is associated with the enhanced expression of integrins ?1?1,
\n ?2?1 and ?3?1 and the acquisition of a migratory phenotype by hepatocytes 56]\u00e2\u20ac\u201c58]. Further, assessment of integrin ?1 expression in human HCC tissues demonstrated
\n a positive correlation with ECM stiffness, pathological grade and metastasis 59]. Blockade of integrin ?1 in vitro<\/em> significantly abrogates migration and invasion of HCC cell lines induced by TGF-?1
\n and epidermal growth factor (EGF) 58], 60]. Conversely, overexpression of integrin ?1 has been reported to enhance HepG2 cell
\n migration 61]. More recently it has been reported that integrin ?1 is involved in the transduction
\n of ECM signalling into HCC cells, resulting in the downstream activation of angiogenic
\n signalling 52]. Utilising a high-stiffness gel to culture HCC cell lines Dong et al.<\/em> found that vascular endothelial growth factor (VEGF) expression is suppressed by
\n treatment with an integrin ?1-specific antibody 52]. SERPINA5 (Protein C inhibitor), a member of the serine protease inhibitor superfamily
\n know to have anti-metastatic and anti-angiogenic effects, 62] is down-regulated in human HCC tissues and further assessment of it\u00e2\u20ac\u2122s anti-tumourigenic
\n activity demonstrated that this was mediated by effects on the fibronectin-integrin
\n ?1 signalling pathway 63]. The relationship between integrin ?1 and ECM stiffness in HCC is further highlighted
\n in a study where resistance of the HCC cell line, Hep3B, to sorafenib was found to
\n be mediated by integrin ?1 and its downstream effector JNK 64].\n <\/p>\n
\n HCC progression. Fan et al.<\/em> have reported integrin ?6 expression to strongly correlate with HCC metastasis in
\n humans 65]. Integrin ?6 overexpression in HCC cell lines (utilising a viral short hairpin RNA-mediated
\n strategy) revealed that integrin ?6 can form a complex with CD151, a tetraspanin protein
\n also associated with HCC invasion 65]. Further investigation in vivo<\/em> indicates that the CD151\/?6 complex stimulates the PI3K-Akt signalling pathway leading
\n to enhanced epithelial-mesenchymal-transition (EMT) of HCC cell lines 65].\n <\/p>\n
\n TGF-? receptor I (TGF-? RI) activation has been reported to promote HCC cell invasiveness
\n through phosphorylation of the intracellular portion of the ?1 subunit of the ?5?1
\n integrin via Smad-2 and Smad-3, leading to an inside-out conformational change and
\n stimulating vascular invasion 66]. Up-regulation of other integrins including ?3?1 and ?6?1 by TGF-?1 has also been
\n reported, leading to increased tumour invasiveness into surrounding tissues 67]. Furthermore specific crosstalk between fibronectin-binding integrins and TGF-?1
\n can promote cell cycle progression in HCC cells through activation of c-Src 68]. Crosstalk between integrins, growth factor receptors and ECM proteins including
\n collagen, have further been shown to alter downstream signal transduction pathways
\n such as Smad, promoting both hepatocyte proliferation and sustaining HSC activation
\n 69], 70]. TGF-?1 has also been reported to modulate ?5?1 expression and synergistically enhance
\n integrin-mediated FAK phosphorylation and cell adhesion in the HCC cell line SMMC-7721
\n 71]. Therefore, integrins (via modulation of TGF-? signalling) may render hepatocytes
\n less sensitive to pro-apoptotic signals in early HCC stages, and more sensitive to
\n tumourigenic differentiation and metastasis formation in advanced HCC.\n <\/p>\nHSC growth factor signalling<\/h4>\n
\n in their secretory phenotype upon activation. In vitro<\/em> studies, using conditioned media from activated HSC, have consistently reported increased
\n proliferation, migration and invasion of tumour cells 72]\u00e2\u20ac\u201c74]. Isolation and subsequent co-culture of human intratumoural HSC with hepatoma cell
\n lines enhanced their viability and migratory capacity 72]. Furthermore, co-transplantation with HCC cells into nude mice promoted tumour formation
\n and growth 75]. Utilising both co-culture and conditioned media from primary human HSC Giannelli
\n and colleagues determined Laminin-5 to be a mediator of HSC-induced HCC migration
\n via its activation of the MEK\/ERK pathway 76]. This is supported by in vivo<\/em> experiments, in which co-transplantation of murine activated HSC with murine HCC
\n cells (H22 line) into immunocompetent mice resulted in significantly larger tumour
\n volumes 73]. Furthermore, implantation of human HCC cell lines (PLC and Hep3B) into nude mice
\n did not form tumours unless activated HSC were concurrently implanted 72]. HepG2 cells did form tumours when implanted alone, however tumour growth was more
\n rapid when co-transplanted with activated HSC 72]. Activated HSC secrete a broad range of growth factors including HGF, TGF-?, fibroblast
\n growth factor (FGF), EGF, VEGF and insulin-like growth factor (IGF). The following
\n sections discuss how these growth factors are involved in HCC pathogenesis.\n <\/p>\nHepatocyte Growth Factor<\/h4>\n
\n survival and angiogenesis 79]\u00e2\u20ac\u201c82]. As such it is widely regarded as a key factor for tumour cell invasion and metastasis
\n 83]. HGF binding to its receptor, c-MET, induces receptor homodimerization and a subsequent
\n phosphorylation cascade. A transmembrane receptor tyrosine kinase, c-MET is found
\n in 20-48\u00c2\u00a0% of HCCs, 84]\u00e2\u20ac\u201c86] and has been shown to be expressed by multiple HCC cell lines 72]. Correlations between increased c-MET and HCC tumour size or invasiveness of HCC
\n have been reported in some studies 87], 88]. c-MET overexpression is also associated with a reduced five-year HCC survival, and
\n a c-MET-regulated expression signature has been reported to define a subset of patients
\n with poor prognosis and an aggressive phenotype 89], 90]. Within HCC tumours, activated HSC have been found to initiate signalling pathways
\n downstream of c-MET, including NF-?B and ERK leading to tumour proliferation and migration
\n 72], 91].\n <\/p>\n
\n was found to abrogate the proliferative and migration-inducing effects on HCC cell
\n lines, seen in non-treated conditioned media 72]. This has also been demonstrated in cancer-associated fibroblasts (CAF) isolated
\n from HCC, where treatment of CAF-conditioned media with an anti-HGF antibody significantly
\n reduced HCC proliferation in Hep3B and MHCC97L cell lines 74]. Moreover, a HGF\/c-MET specific antagonist, NK4, has been found to inhibit markedly
\n the fibroblast-induced invasion of cancer cells, both in vitro<\/em> and in vivo,<\/em>92]\u00e2\u20ac\u201c94] although this has yet to be translated into the clinical setting. A murine model
\n of HCC with similarities to the human disease was recently developed, in which progressive
\n fibrosis and cirrhosis, initiated by ectopic expression of PDGF-C, precedes hepatocyte
\n dysplasia and eventual HCC development 95]. Analysis of these PDGF-C transgenic mice demonstrated that expression of hepatic
\n HGF and its receptor were elevated at the time point at which dysplastic foci are
\n present, further suggesting a pro-tumourigenic role for HGF. Activation of HGF\/c-MET
\n signalling has also been shown to enhance HCC chemoresistance. Conditioned media from
\n the activated HSC cell line LX-2 enhanced resistance of the HCC cell line Hep3B to
\n the chemotherapeutic agent cisplatin, an effect mediated by HGF 96]. Tumour cells may also potentiate pro-metastatic c-MET signalling via an autocrine
\n mechanism involving TIMP-1, leading to downstream expression of metastasis-promoting
\n genes 97], 98].\n <\/p>\n
\n between tumour cells and stromal cells, in particular fibroblasts, has been reported.
\n Nakamura and colleagues have reported the expression of HGF inducers in several carcinoma
\n cell lines, including squamous cell carcinoma, human epidermoid carcinoma, human non-small
\n cell lung cancer cells, human cholangiocarcinoma cells, and SBC-3 human small cell
\n lung carcinoma cells 99]. These HGF inducers include interleukin (IL)-1?, FGF, PDGF and TGF-? and were reported
\n to up-regulate HGF expression by stromal fibroblasts 99], 100]. Taken together, these studies highlight that HGF and aberrant c-MET signalling have
\n a critical role in mediating the bi-directional crosstalk between HSC and tumour cells
\n during hepatocarcinogenesis.\n <\/p>\nTransforming growth factor-?<\/h4>\n
\n and hepatocytes 101], 102] and fixed in the ECM by transglutaminase-dependent linkage of latent TGF-? binding
\n protein to fibronectin and other ECM proteins, forming a reservoir of latent TGF-?.
\n In the context of HCC, it has been suggested that defective TGF-? signalling promotes
\n tumourigenesis secondary to reduced responsiveness to the anti-proliferative effects
\n of TGF-? signalling 103], 104]. However, TGF-? appears to exhibit multiple roles in HCC pathogenesis. Tumour-suppressor
\n functions are observed in the early stages of liver damage and regeneration, whereas
\n during cancer progression, TGF-? may exacerbate tumour invasiveness and metastatic
\n behavior 105]. It has further been demonstrated that TGF-? and PDGF signaling crosstalk supports
\n EMT and is crucial for tumour growth and the acquisition of an invasive phenotype
\n 106].\n <\/p>\n
\n reported to require autocrine TGF-? signalling, with exogenous TGF-? leading to growth
\n inhibition of HCC cells 107]. Utilising HCC cell lines, Meindl-Beinker et al.<\/em> revealed a heterogeneic response to TGF-?, reflective of different stages and mechanisms
\n of disease. Variation between cell lines in their endogenous TGF-? and Smad7 levels,
\n and their transcriptional activity of Smad3, was related to the maintenance of TGF-?
\n cytostatic activity. In particular, the Hep3B, HepG2 and PLC hepatoma cell lines were
\n found to have low TGF-? and Smad7 levels and strong Smad3 transcriptional activity
\n and were thus sensitive to TGF-? cytostatic activity, representative of the early
\n stages of chronic liver disease 108]. In an analysis of TGF-? gene expression in HCC patients, Coulouarn et al.<\/em> reported that those tumours displaying an invasive phenotype and increased recurrence
\n were characterized by a late TGF-? signalling signature, with transcriptional activation
\n of genes associated with matrix remodelling and cell adhesion 109].\n <\/p>\n
\n exhibiting both pro- and anti-tumoural activity, it is highly unlikely that pan-TGF-?
\n blockade will provide a useful therapeutic avenue in HCC treatment. More selective
\n strategies to interfere with TGF-? signalling, perhaps even at a cell-specific level,
\n will likely be required to modulate this signalling pathway for therapeutic gain in
\n the context of HCC.\n <\/p>\nEpiregulin<\/h4>\n
\n cirrhosis, and infectious complications in chronic liver disease. Much interest is
\n currently focused on the translocation of bacterial pathogen-associated molecular
\n patterns (PAMPs), which activate inflammatory responses through Toll-like receptors
\n (TLRs). Recently Dapito et al<\/em>. demonstrated that Tlr4mut<\/em> mice (harbouring non-functional TLR4) that received diethylnitrosamine (DEN) and
\n CCl4<\/sub> show 80-90\u00c2\u00a0% reduction in HCC tumour size and number, compared with mice expressing
\n wild-type TLR4 110]. Gut sterilisation significantly reduced this effect whereas LPS treatment enhanced
\n it, suggesting a role for the LPS-TLR4 pathway in promotion of hepatocarcinogenesis.
\n Interestingly, alongside hepatocytes, HSC were identified as candidates for TLR4-dependent
\n tumour promotion in the chronically injured liver. LPS and the gut microbiome were
\n found to induce HSC activation, resulting in production of the mitogens HGF and epiregulin,
\n which likely act on malignant hepatocytes. Epiregulin is a member of the EGF family,
\n and results in EGF receptor and human epidermal growth factor receptor 2 activation
\n during early stages of DEN\/CCl4<\/sub> carcinogenesis, whereas it reduces hepatocyte apoptosis by NF-KB nuclear translocation
\n during later stages 110], 111]. This suggests that there may be merit in evaluating whether long-term antibiotic
\n treatment confers any protection against HCC development. This could initially be
\n investigated by following up patients with cirrhosis on long-term prophylaxis for
\n spontaneous bacterial peritonitis or encephalopathy, although identifying a comparable
\n control group may prove challenging.\n <\/p>\nHSC and angiogenesis<\/h4>\n
\n reflected by the efficacy of sorafenib, which targets this process. The rapid growth
\n pattern of malignant hepatocytes requires new vessel formation, stimulated by multiple
\n pro-angiogenic factors. This pro-angiogenic environment in turn supports tumour progression
\n and metastasis. The relevance of tumour vascularity is reinforced by the observation
\n that VEGF expression progressively increases from low-grade dysplasia to early-stage
\n HCC 112]. VEGF overexpression is also associated with high tumour grade, and vascular and
\n portal vein invasion 113]\u00e2\u20ac\u201c117]. Furthermore, raised plasma VEGF and angiopoietin 2 (Ang-2) are independent predictors
\n of poor prognosis in advanced HCC 118].\n <\/p>\n
\n MMPs, FGF, TGF-?1, EGF, angiopoietin-1 (Ang-1) and Ang-2 119]\u00e2\u20ac\u201c121]. Upon activation, HSC express multiple smooth muscle cell markers, suggesting they
\n may act like pericytes during angiogenesis 122], 123]. They also express angiogenic growth factor receptors, such as VEGF receptor, PDGF
\n receptor and Tie-2 124]\u00e2\u20ac\u201c126]. In liver injury and HCC, this facilitates reciprocal signalling between HSC and
\n endothelial cells or malignant hepatocytes and contributes towards a pro-angiogenic
\n microenvironment. VEGF secretion by HSC can be hormonally induced by leptin, or by
\n physical stress such as hypoxia, and is upregulated in HCC 120], 124], 127]. VEGF receptor upregulation also occurs during HSC activation, resulting in increased
\n mitogenesis in response to VEGF 13].\n <\/p>\n
et al.<\/em> co-cultured LX2 cells with HepRG HCC cells, and analysis of differential gene expression
\n identified a gene network linked to VEGFA and MMP9 128]. This was shown to promote angiogenesis, as conditioned medium from LX2-HepaRG coculture
\n (but not LX2 or HepaRG medium alone) induced tubule complex formation by primary human
\n umbilical vein endothelial cells. A gene signature of this cross-talk correlated with
\n poor prognosis and metastasis in humans 128].\n <\/p>\n
\n 129]. They went on to demonstrate in vivo<\/em>, using an orthotopic HCC model, that activated HSCs promote tumour vascularisation
\n via increased VEGF and possibly PDGF secretion.\n <\/p>\n
\n cells and activated HSC. Torimura et al.<\/em> characterised expression of Ang-1, Ang-2 and Tie2 receptors in HCC cell lines (HLE
\n and HuH-7) and human HCC cases 130]. They concluded angiopoietin-Tie2 signalling in the vascular wall may act in favour
\n of vessel remodelling in HCC. Ang-2 production by hepatoma cells, HSC and smooth muscle
\n cells binds Tie2 (on HSC, smooth muscle and endothelial cells) and destabilises connections
\n between endothelial cells, perivascular support cells and ECM. This allows exposure
\n to VEGF, which in these relatively hypoxic conditions, is upregulated. Proliferation
\n of endothelial cells ensues, allowing neovascularization and further tumour growth.\n <\/p>\n
\n was acting via AMPK activation, and specifically targeting HSC in this model. Indeed,
\n inhibition of AMPK on LX2 cells (but not on HepG2 cells) using siRNA did restore VEGF
\n levels and abrogate metformin\u00e2\u20ac\u2122s anti-angiogenic effect. Metformin would seem a promising
\n candidate for human HCC treatment, but unfortunately retrospective data would suggest
\n a lack of survival benefit 132]. However, considering the well-established tolerability of metformin, its potential
\n HSC-mediated effect on angiogenesis merits further investigation.\n <\/p>\n
\n Similarly, HCC signalling results in further HGF production from activated HSC. TGF-?
\n demonstrates both tumour-suppressive and tumour-promoting functions, depending on
\n context. HSC produce angiogenic cytokines, supporting new vessel growth. HCC cells
\n contribute to angiogenic signalling, and HSC also possess receptors for some of these
\n factors. Gut-derived LPS induces HSC activation, resulting in epiregulin and HGF production,
\n with mitogenic effects on HCC\n <\/p>\nHSC and immunomodulation<\/h4>\n
\n a very active area of research. One mechanism by which tumours evade the immune response
\n is through the augmentation of the numbers and activity of immunosuppressive cells,
\n at both the tumour site and within lymphoid organs 133]. Such cells include regulatory T-cells (Tregs) and myeloid-derived suppressor cells
\n (MDSC). Increased levels of Tregs within peripheral blood and tumours have been reported
\n in human HCC cases, and have further been shown to suppress anti-tumour immune responses
\n in addition to promoting angiogenic remodeling 134]\u00e2\u20ac\u201c136]. Further, intratumoural Treg accumulation has been reported to correlate with disease
\n progression and poor prognosis 137]. MDSC are defined by the markers CD11b and Ly6-C\/G and have been found in the tumour,
\n lymph nodes and blood, suppressing cellular responses to cancer cells 138].\n <\/p>\n
\n demonstrating, both in vitro<\/em> and in vivo<\/em>, that activated HSC are able to strongly suppress T-cell responses. Investigation
\n into the divergent immunomodulatory activity of quiescent and intratumoural HSC has
\n revealed that, in vitro<\/em>, intratumoural HSC induce T-cell hyporesponsiveness, an effect not seen with quiescent
\n HSC 139]. Moreover, in an orthotopic rat model of HCC, intratumoural HSC number strongly correlated
\n with T-cell apoptosis and lung metastatic nodules 140]. Although a direct interaction was not reported, this does suggest an additional
\n role for HSC in HCC metastasis via an immunosuppressive mechanism.\n <\/p>\n
\n and enhanced tumour angiogenesis, in association with inhibition of lymphocyte infiltration
\n and apoptosis of infiltrating monocytes 73]. In an orthotopic model of HCC, activated HSC in tumour-bearing mice significantly
\n increase Treg and MDSC populations in the spleen and tumour stroma 141]. An increase in tumour vascular and lymphatic vessel density was also reported in
\n those tumours co-transplanted with HSC.\n <\/p>\n
\n demonstrated that this may be mediated via upregulation of human B7 homolog 1 (B7-H1;
\n programmed death ligand 1 (PDL-1)) on tumoural HSC 142]\u00e2\u20ac\u201c144]. B7-H1 can act as both receptor and ligand and has immunosuppressive functions such
\n as promoting activated T-cell apoptosis and inhibiting T-cell-mediated tumour cell
\n apoptosis 1], 145], 146]. Its counter-receptor, PD-1, is expressed on activated, but not resting, T-cells,
\n B-cells and monocytes 2]. B7-H1\/PD-1 signaling has been reported to promote Treg cell induction and immunosuppressive
\n function through the down-regulation of mTOR and AKT phosphorylation 147], 148]. In vitro<\/em> experiments involving incubation of T-cells with anti-B7-H1 monoclonal antibody resulted
\n in a significant reduction in HSC immunomodulatory activity and HCC migration and
\n invasion 139].\n <\/p>\n
\n and promising Phase 1 data has been reported 149]. In one study, varying degrees of tumour regression were found in colon, renal and
\n lung cancers and melanoma and a significant increase in tumour lymphocyte infiltration
\n was noted 150]. This has been extended to a second clinical trial where responses were seen in 16
\n out of 39 patients with advanced melanoma 151]. These early clinical studies further demonstrated encouraging safety data. In the
\n context of HCC, a Phase 1, dose escalation study investigating the effects of anti-PD-1
\n therapy is currently underway in patients with advanced HCC (NCT01658878), however
\n results have yet to be reported. Some of the immunomodulatory effects of HSC in HCC
\n are summarised in Fig.\u00c2\u00a02.<\/p>\n
\n through i) an increase in Treg cell induction and immunosuppressive function and ii)
\n upregulation of B7-H1 on iHSC resulting in increased ligation of its receptor (PD-1)
\n on activated T-cells, leading to increased apoptosis of activated T cells with subsequent
\n inhibition of T-cell-mediated tumour cell apoptosis. This results in HCC immunotolerance
\n and a permissive environment for tumour growth. PD-1, programmed death ligand; B7-H1
\n human B7 homolog 1; ECM, extracellular matrix\n <\/p>\nTherapeutic approaches to targeting HSC and HSC signalling<\/h4>\n
\n targeting of HSC remains challenging. Recently, transgenic mice have been developed
\n that allow reliable fluorescent labeling or genetic manipulation in HSC and myofibroblasts
\n 152], 153]. These transgenic mice will hopefully prove useful not only in elucidating the molecular
\n mechanisms in HSC that regulate the stroma-HCC interface, but also in facilitating
\n the identification of rational, new therapeutic targets in hepatocarcinogenesis.\n <\/p>\n
\n cell-specific therapy is conceivable, albeit not entirely straightforward. ECM homeostasis
\n is a key physiological process and modifying HSC functions may impair this, with potential
\n for severe adverse effects. Practically, delivering drugs to HSC is hindered by a
\n lack of multiple transport receptors and endocytic capacity. Furthermore, candidate
\n compounds may include siRNA and cytokines, which have a short half-life in plasma
\n following systemic administration, hindering therapeutic efficacy 154].\n <\/p>\n
\n to deliver therapeutic compounds. This involves coupling the selected compound to
\n a carrier possessing a specific receptor-binding ligand, or an antibody.\n <\/p>\n
\n on HSC, and a liposome specific to the vitamin A receptor on HSC 155], 156]. Furthermore Poelstra et al.<\/em> have used proteins substituted with a sugar moiety that binds the mannose-6-phosphate-IGFII
\n receptor 157]. They have also utilised a peptide that binds the PDGF receptor-?, 158] to deliver a protein or an adenovirus to HSC 159], 160]. An RGD-peptide which binds to RGD-binding integrins has also been used to create
\n a carrier that accumulates in HSC 161], 162]. Of note, the carrier molecules used must fit strict criteria such as low immunogenicity,
\n and high stability, biocompatibility and selectivity, if they are to translate into
\n clinical practice. Moreover, the target receptors on HSC should be selectively expressed
\n and ideally upregulated during disease activity. A further challenge is the requirement
\n for endocytosis of the construct following target receptor binding. This can be particularly
\n problematic in the case of biological therapeutics, which usually fail to withstand
\n the endosomal degradation process.\n <\/p>\n
\n (IFN?) to HSC 163]. This elegant system transported the signalling moiety of IFN? to the PDGF-receptor
\n with a carrier molecule that was simplified and miniaturised. They found that IFN?
\n could be effectively delivered to human HSC in vitro<\/em>, and to mouse HSC in vivo<\/em>. Furthermore, the targeted fusion proteins were shown to ameliorate hepatic fibrosis
\n in CCl4<\/sub>-treated mice 163]\u00e2\u20ac\u201c165]. This suggests that directing a cytokine to HSC is a feasible and potentially tractable
\n therapeutic approach, both in the context of developing new treatments for patients
\n with liver fibrosis, as well as HCC. Therapeutic approaches to targeting HSC are summarised
\n in Fig.\u00c2\u00a03.<\/p>\n
\n to a carrier possessing either a HSC-specific receptor-binding ligand or an antibody.
\n Carriers utilised include: a monoclonal human single chain antibody (scAb) fragment
\n to synaptophysin 155]; a sugar moiety that binds the mannose-6-phosphate (M6P) insulin-like growth factor
\n receptor 157]; a liposome specific to the vitamin A (retinol-binding protein) receptor 156]; PDGF?-peptide 160]; PDGF? receptor recognising peptide (PPB) 164]; an RGD peptide bound to a liposome or coupled to human serum albumin (HSA) 159], 162] scAb Fv, single chain antibody variable fragment; PEG, polyethylene glycol; pCVI,
\n 10 cyclic peptide moieties that recognise collagen type VI receptors\n <\/p>\n","protected":false},"excerpt":{"rendered":"