Angiogenesis, the formation of new vascular structures from preexisting vessels, occurs in several organs during multiple pathophysiological situations. Although traditionally associated with tumorigenic processes, hepatic angiogenesis has also been observed in the context of different inflammatory, fibrotic, and ischemic conditions. It is unclear whether angiogenesis merely represents a homeostatic mechanism aimed at ensuring an adequate oxygen supply or one that exerts an additional pathogenetic role contributing to liver damage.1, 2 This latter hypothesis has recently generated great interest, because, if confirmed, angiogenesis could be used as a potential prognostic marker of disease progression and as a novel therapeutic target for these disorders. In the present article, we review the evidence for angiogenesis in chronic inflammatory liver diseases and analyze the cellular and molecular mechanisms involved. Angiogenesis in hepatocellular carcinoma is beyond the scope of this short review, deserving an independent analysis in depth.
Intrahepatic hypoxia may occur during the inflammatory and fibrotic processes that characterize several chronic liver diseases of viral and autoimmune origin. As a consequence, new vascular structures are formed to provide oxygen and nutrients. Angiogenesis involves a tightly regulated network of cellular and molecular mechanisms that result in the formation of functional vessels. Of particular importance are growth factors, molecules involved in matrix remodeling and cell migration, and vessel maturation—related factors. In recent years, a number of studies have examined the expression and function of many pro- and antiangiogenic molecules in the setting of nontumoral chronic liver diseases and liver regeneration. This review examines the potential pathogenetic role of angiogenesis in the context of viral hepatitis, cirrhosis, autoimmune hepatitis, primary biliary cirrhosis, and alcoholic liver disease. The future perspectives for research in this field are outlined. (HEPATOLOGY 2004;39:1185–1195.)
Molecular Insights Into the Angiogenic Process
Hypoxia is the main stimulus for angiogenesis by signaling through hypoxia-inducible transcription factors.3 The formation of new functional vessels involves different processes whose molecular effectors must be precisely regulated1, 2 (Fig. 1 and Table 1).
|VEGF family members||Increase vascular permeability; induce EC proliferation; leukocyte adhesion; regulate neovessel lumen diameter|
|VEGF-R, NRP-1||Integrate angiogenic and survival signals|
|Integrins αvβ3, αvβ5, α6β1||ECM receptors, intercellular communication; mobilized during EC migration; regulate neovessel lumen diameter|
|uPA||Remodels ECM; releases and activates growth factors|
|PAI-1||Stabilizes nascent vessels|
|MMPs, heparinases, chymases, tryptases, cathepsins||Remodel ECM; release and activate growth factors|
|PlGF, aFGF, bFGF, HGF, TGF-α, TGF-β||Induce EC proliferation|
|MCP-1 and other chemokines||Pleiotropic role in angiogenesis|
|MEF2C||Regulates neovessel lumen diameter|
|Ephrins||Determine branching and arterial/venous specification|
|PDGF-B and receptors||Recruit pericytes|
|Ang-1||Stabilizes intercellular contacts; inhibits permeability|
|Tie-2||Receptor for Ang-1 and Ang-2|
|Ang-2||Ang-1 antagonist (destabilizes vessels; causes EC death)|
|TGF-β1, endoglin||Promote vessel maturation, stimulate ECM generation, induce differentiation of mesenchymal cells to pericytes|
|Cyr61, Fisp12||Stimulate directed migration of EC through an αvβ3 integrin-dependent pathway; ECM modifiers, promote EC survival|
|VE-cadherin, PECAM-1, plakoglobin, β-catenin||Adherens junction molecules; intercellular adhesion; provide vessel tightness|
|Claudins, occludin, JAM-1, -2, -3||Tight junction molecules; intercellular adhesion, provide vessel tightness|
|Connexins||Gap junction molecules; facilitate intercellular communication|
|Integrins αvβ3, αvβ5||Suppress VEGF- and Flk-1-mediated EC survival|
|TIMPs||Inhibit ECM degradation by MMPs; inhibit EC proliferation|
|Angiostatin and related plasminogen fragments||Suppress tumor angiogenesis|
|Endostatin, antithrombin III, IFN-β, LIF, PF4||Suppress EC cell proliferation|
|TSP-1||Inhibits lumen formation|
|Ang-1 (excess)||Makes vessels too tight and inhibits sprouting|
|Ang-2||Facilitates sprouting in the presence of VEGF|
Endothelial budding is facilitated by vasodilation, loosening of interendothelial contacts, and leakiness of preexisting vessels, which allows extravasation of plasma proteins (fibrinogen, fibrin) that, together with extracellular matrix components (ECM), lay down a provisional scaffold for migrating endothelial cells (ECs). Nitric oxide (NO), whose angiogenic properties have been characterized,4 is the main factor responsible for vasodilation, whereas vascular endothelial growth factor (VEGF) increases vascular permeability. These factors must overcome the forces that provide mechanical strength and tightness to the vessels, comprising (1) adherens junctions established by vascular endothelial (VE) cadherin, platelet endothelial cell adhesion molecule-1 (or CD31), and their corresponding cytoskeleton-linking molecules (plakoglobin, β-catenin); (2) tight junctions formed by claudins, occludin, junctional adhesion molecules 1, 2, and 3; and (3) gap junctions mediating intercellular communication through connexins. Whereas VEGF loosens these contacts, angiopoietin-1 (Ang-1) stabilizes them through mechanisms not fully understood. In parallel, mobilization of integrin/tetraspanin complexes to emerging motile-associated structures must occur to facilitate EC migration.1, 2
ECM Remodeling and EC Migration.
Next, the basement membrane (mainly collagen IV and laminin) and the ECM (collagen I, elastin) must be degraded to allow subsequent EC migration and proliferation. This is performed by specialized proteinases, including plasminogen activators—urokinase plasminogen activator (uPA) and plasminogen activator inhibitor-1 (PAI-1), matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs), heparinases, chymases, tryptases, and cathepsins.5, 6 ECM proteolysis leads to the exposure of cryptic epitopes and release of ECM-embedded factors that promote EC migration and proliferation. Integrins participate in EC migration by regulating detachment/attachment to the ECM and maintaining communication between the EC and its neighborhood.7 Integrins αvβ3 and αvβ5 have long been considered proangiogenic, but recent data indicate that they may inhibit angiogenesis by suppressing VEGF- and Flk-1-mediated EC survival.7 Insufficient or inadequate proteolysis precludes EC migration,8–10 whereas excessive ECM degradation destabilizes the support structure on which ECs migrate, thereby inhibiting angiogenesis in both cases.11 The activity of proteinases is context- and concentration-dependent, to allow a tight regulation of ECM remodeling. For instance, the activity of membrane type-1 MMP (MT1-MMP) is spatially regulated at the subcellular level by its differential association to β1 or β3 integrins.10 Furthermore, the formation of integrin/tetraspanin complexes could regulate the localization of MMPs to motile structures of migrating ECs, as occurs in keratinocytes or epithelial cells.12, 13
Once the path is cleared, ECs proliferate in response to growth factors secreted by ECs or surrounding cells (including hepatic stellate cells—considered liver-resident pericytes—leukocytes, hepatocytes, and Kupffer cells) and assemble into new capillary tubes. The most thoroughly characterized is VEGF, a multifunctional protein that binds to 2 tyrosine kinase receptors: kinase insert domain receptor (KDR) and fms-like tyrosine kinase receptor (Flt-1).14 The VEGF promoter contains hypoxia-inducible factors–responsive elements. VEGF plays a crucial role in virtually all pathological situations in which angiogenesis occurs, and the therapeutic use of strategies aimed at blocking its mode of action is currently under evaluation. EC proliferation may also be stimulated by additional members of the VEGF family: placental growth factor, which potentiates the effects of VEGF and plays a relevant role during angiogenesis; VEGF-C and VEGF-D, which regulate lymphatic angiogenesis; and VEGF-B, recently described to promote in vivo angiogenesis, particularly in the setting of inflammatory arthritis.15, 16 In addition, EC proliferation may be stimulated by other growth factors, such as acidic and basic fibroblast growth factors (aFGF and bFGF); hepatocyte growth factor (HGF); and transforming growth factor (TGF) α and β, whose pathophysiological relevance is less well characterized.2, 17 Cytokines, lipid mediators, hormones, and neuropeptides positively regulate EC growth as well, whereas molecules such as angiostatin (an internal fragment of plasminogen), endostatin (a fragment of collagen XVIII), antithrombin III, interferon β, leukemia inhibitory factor, and platelet factor 4 suppress EC proliferation.2, 17 Different chemokines have also been suggested to be pro- or antiangiogenic for ECs.18, 19
ECs proliferate in an ordered manner that leads to the formation of a lumen, whose diameter and length is influenced by VEGF, Ang-1, integrins αvβ3 and αv5 and the myocyte enhancer-binding factor 2C transcription factor, whereas thrombospondin inhibits lumen formation.20, 21 If excessive proteolysis occurs, ECs assemble in cysts rather than in tubular structures due to the degradation of the scaffold and cues required for adequate migration. A structured 3-dimensional network of vessels of uniform size is then organized by carefully regulated mechanisms involving signaling pathways that determine branching (such as ephrins and neuropilins), formation of basement membrane and ECM components (regulated by MMPs and TIMPs), and cell migration and differentiation.
Stabilization of Nascent Vessels.
For nascent vessels to mature, pericytes must be recruited, and a new basement membrane and ECM must be generated to provide structural stabilization.21 Physical forces and multiple molecules contribute to these processes. By releasing platelet-derived growth factor B (PDGF-B), ECs attract and stimulate proliferation of PDGF-Rβ-expressing pericytes.22 Pericyte-secreted Ang-1 stabilizes nascent vessels by binding the Tie-2 receptor, thereby affecting junctional molecules23 and facilitating communication between ECs and mural cells.24 However, an excess of Ang-1 makes vessels too tight and inhibits sprouting.25 Ang-2 may exert opposing effects: in the absence of VEGF, Ang-2 acts as an antagonist of Ang-1, destabilizes vessels, and causes EC death, leading to vessel regression26; however, it facilitates sprouting in the presence of VEGF.25 Finally, TGF-β1 promotes vessel maturation by stimulating generation of ECM and inducing differentiation of mesenchymal cells to pericytes. The TGF-β1-ALK5 pathway positively regulates vessel maturation.27
Most molecular mechanisms of angiogenesis are common to the liver and other organs. However, potential differences might arise from (1) the existence of 2 types of microvascular structures in the liver—large vessels, such as portal and central venules and hepatic arterioles, which are lined with continuous EC, and sinusoids, which are lined with sinusoidal ECs that are discontinuous and present fenestrations28; (2) the production of unique proangiogenic factors, such as the recently described ANGPTL3, a liver-specific secreted factor showing angiogenic properties by binding to αvβ3 integrin29; and (3) the fact that hepatic stellate cells, considered liver-resident pericytes, might contribute to angiogenesis through mechanisms different from those attributed to microcapillary pericytes. This issue is currently under study. Physiological hepatic angiogenesis occurs during liver regeneration, leading to the formation of new functional sinusoids, whereas pathological angiogenesis occurs in fibrosis and is characterized by the appearance of capillarized vascular structures.
Liver Regeneration: A Model for Hepatic Angiogenesis
Much of the current knowledge on physiological hepatic angiogenesis stems from investigations on liver regeneration following acute liver damage or partial hepatectomy (PH),30–32 the latter being considered an appropriate model to investigate physiological angiogenesis.33–35 Some of the molecular mechanisms involved in hepatic tissue revascularization have been recently unveiled.30–32, 35–38
Following PH in rodents, hepatocyte proliferation starts, with maximum activity in periportal areas39; reconstruction of sinusoids by ECs begins from these periportal areas.36 Sinusoidal EC proliferation and upregulation of hypoxia-inducible genes occurs at a later stage in response to hypoxia generated within the initially formed avascular clusters of hepatocytes. The endothelial expression (and activation) of several proangiogenic growth factor receptors has been recently characterized.35 Most of them are expressed at very low levels in resting liver ECs and are significantly induced and tyrosine-phosphorylated (reflecting activation) during liver regeneration (Table 2).
|Growth Factor/Receptor||Expressing Cells/Structures||Increase (Post-PH Period)|
|VEGF40, 41, 43||Periportal hepatocytes, HSC||48–72 h|
|Flt-135, 37, 43||Arterioles and sinusoidal EC, HSC||72 h to 12 d|
|Flk-1/KDR35||Large vessels and sinusoidal EC||72 h to 10 d|
|Angiopoietin-137||Unknown||72–96 h (mRNA)|
|Angiopoietin-237||Unknown||72–168 h (mRNA)|
|Tie-235||Sinusoidal and large vessel EC||Start at 48 h, peak at 96 h|
|Tie-135||EC around avascular hepatic islands||Start at 48 h, peak at 96 h|
|PDGF93, 114||Nonparenchymal cells, including EC and HSC||48 h post-liver injury (CCl4)|
|PDGF-Rβ35||Sinusoidal and large vessel EC, HSC||3–12 d|
|aFGF115||Stellate cells||24 h (mRNA)|
|Flg (FGF-R1)35, 116||Hepatocytes (low levels)||Unchanged|
|Bek (FGF-R2)35, 116||Hepatocytes (very low expression)||Not determined|
|EGF31||Hepatocytes, extrahepatic sources||Immediate, peak 12–24 h|
|TGF-α31, 117||Hepatocytes||12–24 h (mRNA)|
|EGF-R35||Sinusoidal EC; large vessels (minor)||Constitutive, 3–14 d|
|HGF118||Nonparenchymal cells, including EC||Immediate, peak 12–24 h|
|c-Met35||EC, hepatocytes, biliary epithelium||No change in EC|
VEGF is upregulated in periportal hepatocytes shortly after PH,36, 40 and aside from its autocrine action to induce hepatocellular growth, it stimulates surrounding ECs in a paracrine fashion.41 Its involvement in liver regeneration-associated angiogenesis has been shown by (1) the inhibition of hepatocellular and sinusoidal EC proliferation achieved by injecting rats with a mAb anti-VEGF at the time of PH, and (2) the significant potentiation of hepatocellular and sinusoidal EC proliferation by exogenous VEGF injection simultaneously with PH.36 However, the role of VEGF during liver regeneration may be broader than the simple stimulation of cell proliferation: selective activation of VEGF receptor Flt-1 primes liver sinusoidal ECs to produce a series of mitogenic/survival factors that can protect parenchymal cells from injury and initiate regeneration.42 VEGF, along with its receptor Flt-1, can also be expressed by activated hepatic stellate cells in response to hypoxic stimuli.43 VEGF effects are not confined to sinusoidal ECs and hepatocytes because hepatic stellate cells can also proliferate in response to this growth factor.44, 45 Hepatic expression of Ang-1 and Ang-2 is increased following PH in rats, but at a later phase than VEGF.37 Both factors exert opposing actions: in the presence of VEGF, Ang-2 seems to augment angiogenesis early during regeneration, while in the absence of VEGF, Ang-2 inhibits vascular growth at later stages.37
The profile of activation of PDGF and its receptors during liver regeneration (Table 1) suggests a potential role in maturation of the newly formed sinusoids.35 FGF receptors FGF-R1 (Flg) and FGF-R2 (Bek) are increased in hepatocytes during liver regeneration following PH,46 but there is limited evidence of their contribution to vascular growth.35 However, recent data show that injection of exogenous bFGF at the time of PH increases EC proliferation and blocks apoptosis, thus accelerating hepatic regeneration.47
Growth factor receptors with a different pattern of induction during liver regeneration include EGF-R and the HGF receptor c-Met. EGF-R is highly expressed under healthy conditions (suggesting a homeostatic function in the maintenance of vascular structures) and contributes to angiogenesis during liver regeneration by playing a supportive role. HGF receptor c-Met, expressed only in large vessels, shows no change during regeneration. However, the hypoxia-induced activation of the HGF/c-Met pathway is essential for liver regeneration following PH in rats: HGF is not upregulated and c-Met is inhibited following PH in diethylnitrosamine-induced experimental cirrhotic rats, resulting in failure of liver regeneration.48 These data suggest that the critical role of HGF/c-Met during liver regeneration affects hepatocellular or biliary epithelial mitogenesis more importantly than microvascular remodeling.35
Differential gene expression analysis of mouse livers immediately after PH (i.e., during the “priming phase” of liver regeneration as defined by Fausto32) showed a marked upregulation of proangiogenic ECM modifiers Cyr61 and Fisp12 as early as 10 minutes and 90 minutes, respectively, after PH.49 These genes are critical for neovessel formation by the stimulation of directed migration of human microvascular ECs through an αvβ3 integrin-dependent pathway.50, 51 In contrast, little or no induction of MMPs or TIMPs was detected up to 4 hours following PH. Other studies, however, showed overexpression of specific MMPs at later time points.52–56 Expression of TIMP-1 in mesenchymal cells is significantly increased 24 hours after PH in rats. TIMP-1 contributes to regulation of the increased MMP-2 and MMP-9 activities secondary to pro–MMP-2 and pro–MMP-9 induction. MT1-MMP is also upregulated, with potential implications for pericellular fibrinolysis, as is required both for hepatocellular replication and for EC migration and angiogenesis. Finally, uPA plays a role in post-PH-related angiogenesis in mice because pharmacological uPA inhibition leads to delay and incomplete restoration of sinusoidal architecture. During the priming phase of liver regeneration in mice, several protooncogenes, as well as a number of transcription factors involved in the elicitation of inflammation and angiogenesis (including hypoxia-inducible factor-1α), were also specifically upregulated.30, 32, 49, 57–60
Angiogenesis in Chronic Inflammatory Liver Injury
Most chronic liver diseases are characterized by fibrosis and inflammation. During the fibrogenic process, an excessive amount of ECM is synthesized and accumulated.61 Fibrotic tissue offers resistance to blood flow and to the delivery of oxygen, thus becoming hypoxic. Stimulation of hypoxia-inducible factors leads to an angiogenic switch, to the upregulation of proangiogenic factors, and to the formation of neovessels62 (Fig. 2). Consistently, several antiangiogenic therapeutic strategies have been shown to suppress liver fibrosis development in experimental animals. These include TNP-470 (a proangiogenic semisynthetic analog of fumagilin),63 vascular endothelial growth factor receptor 1 and 2-neutralizing monoclonal antibodies,64 and angiotensin I-converting enzyme inhibitors (angiotensin II is proangiogenic).65 Further therapeutic approaches targeting activation of hepatic stellate cells and regulation of ECM degradation might show beneficial effects in fibrosis, at least in part, by interfering with the development of angiogenesis. This hypothesis requires further investigation.
Similarly, a local inflammatory reaction exists in most liver diseases. Proinflammatory mediators, as well as other nonhypoxic stimuli, can elicit an angiogenic response through the induction of hypoxia-inducible factor-1α and HIF-1-dependent transcriptional activity, including VEGF production.66 In addition, the production of proinflammatory eicosanoids by cyclooxygenase-2 plays a role in neovasculature formation because its inhibition by selective antagonists blocks angiogenesis, both in experimental animals and in humans.67 The mechanism of action is not fully understood, but it involves reduction of proliferation and induction of apoptosis in angiogenic ECs.68 The therapeutic potential of selective cyclooxygenase-2 inhibitors has been shown mainly in tumor-related angiogenesis.69 However, these may also be effective in nontumoral hepatic angiogenesis because cyclooxygenase-2 activation occurs in chronic liver inflammation.70 During inflammation, vascular permeability is increased and monocytes, macrophages, platelets, mast cells, and other leukocytes (able to produce angiogenic cytokines and growth factors) are recruited71 under the attraction of chemokines.19, 72 As for other organs (bone, joints, lung, and skin), fibrosis and/or inflammation contribute to the formation of new vascular structures in the liver.73 The currently available information on the occurrence of angiogenesis in different chronic inflammatory liver diseases is reviewed.
Chronic Viral Hepatitis.
Angiogenesis occurs in the liver of patients infected with hepatitis B or C virus. The appearance of ECs forming characteristic capillary structures in inflamed portal tracts from chronic viral hepatitis has been demonstrated.62, 74 The pathophysiological significance of chronic viral hepatitis-associated angiogenesis is presently unclear; it has been proposed to exert a beneficial role by contributing to tissue repair and regeneration after liver damage.62 It has also been suggested to represent a risk factor for progression to hepatocellular carcinoma in patients with chronic hepatitis C.75
The molecular mechanisms involved in chronic viral hepatitis-associated angiogenesis have not been fully identified. However, many of the molecules that have been shown either in vitro or in experimental animals to participate in the angiogenic response are known to be overexpressed in the livers of such patients. This suggests their potential role in the development of angiogenesis, although causation has not been demonstrated in all cases. Local production of NO as a result of the overexpression of inducible NO synthase in the livers of patients infected with hepatitis C or B virus may participate in the angiogenic response by inducing vasodilation.76, 77 VEGF and HGF, whose expression is increased during chronic viral hepatitis,78–80 may contribute to enhancing vascular permeability, as suggested by the effects described for these 2 growth factors. VEGF has been reported to induce NO-mediated vasodilation,14 most probably contributing to the progression of neoangiogenesis.62 HGF effects include a decrease and redistribution of VE-cadherin (which participates in intercellular contacts), β-catenin, and plakoglobin (linking molecules between VE-cadherin and the actin cytoskeleton) in ECs. Additionally, viral proteins may also play a role in inducing a disruption of interendothelial junctions81 through mechanisms involving Src kinases, molecules required for vascular permeability during angiogenesis.82 Binding between ECs and the ECM may be altered in the livers of chronic viral hepatitis patients.83 Integrin αvβ3 (expressed by activated ECs), shows increased tissular expression in chronic hepatitis C.80 Interestingly, when ECs are induced to migrate by HGF, αvβ3 integrin accumulates at the leading edge,80 which represents the anchoring zone required for pulling the cellular body.84
MMPs and their inhibitors may also be involved in angiogenesis during chronic viral hepatitis, as can be deduced from their mode of action in experimental models: they contribute to migration and proliferation of activated ECs by regulating peripheral ECM degradation and remodeling (effects that may also be of relevance in fibrogenesis and/or tumor invasion85, 86). MMP induction in the liver is mainly caused by proinflammatory and profibrogenic factors, although the virus may also stimulate their expression: infection of hepatocytes and HepG2 cells with hepatitis B virus affects the upregulation of MMP-2 expression and activation, thereby increasing the invasion potential of transformed cells by plasmin.87 Furthermore, infection of hepatocytes with the hepatitis B virus X protein induces an upregulation of MT1-MMP expression, which in turn activates MMP-2, thus resulting in an enhancement of hepatocyte invasion both in vivo and in vitro. Induction of both MT1-MMP expression and cell invasion by the hepatitis B virus X protein is dependent on cyclooxygenase-2 activity.
Hepatic VEGF messenger RNA78 and protein80 levels are increased in chronic hepatitis C patients, suggesting a potential role in the induction of EC proliferation in this disease, although a direct causal relationship has not been demonstrated. Similarly, the hepatic expression of HGF, another potent inducer of angiogenesis both in vitro and in vivo,88–90 is also upregulated during chronic viral hepatitis79 and other chronic inflammatory liver diseases.91 HGF induces EC proliferation in a VEGF-dependent manner.80, 90 The expression of PDGF in macrophages and infiltrating inflammatory cells and the expression of PDGF receptors in sinusoidal and perisinusoidal cells in periportal areas are increased during chronic viral hepatitis and/or cirrhosis,92, 93 suggesting a role in the stabilization of nascent vessels1, 94 during chronic viral hepatitis-associated angiogenesis.
An enhanced hepatic vascular proliferation or “vascular remodeling,” consistent at least in part with the occurrence of pathological angiogenesis, has been described in association with liver cirrhosis.95, 96 Angiogenesis has also been observed in experimental biliary cirrhosis: following administration of diethylnitrosamine to rats, the progression of liver fibrosis was seen to be associated with hepatocellular hypoxia and angiogenesis, and hepatic VEGF and Flt-1 expressions were increased and correlated with microvessel density. The activation of hepatic stellate cells played a central role in this setting, through the enhanced expression and secretion of both proangiogenic and fibrogenic factors.97, 98
DNA array analysis of differential gene expression in human cirrhotic livers from patients with chronic hepatitis C, autoimmune hepatitis, primary biliary cirrhosis, and primary sclerosing cholangitis showed overexpression of many of the key genes involved in the different phases of angiogenesis compared to nondiseased liver tissue: growth factors and their receptors (e.g., VEGF, HGF, FGF-8, FGF receptor-1, PDGF), cell-cell and cell-matrix adhesion molecules (integrins, beta-catenin), matrix remodeling molecules (extracellular MMP inducer), PAI-3), molecules involved in vascular differentiation and polarity (ephrins) and many others.99, 100 The precise mechanisms by which these genes contribute to the formation of neovessels in liver cirrhosis remain undetermined.
Autoimmune Liver Diseases.
Controversy exists regarding intrahepatic vasculature in autoimmune liver diseases. Some authors have reported a tendency to vasopenia and decreased peribiliary capillary plexus in the livers of primary biliary cirrhosis, primary sclerosing cholangitis, and autoimmune hepatitis patients.101, 102 This has been attributed to destruction of vascular structures by autoimmune mechanisms, similar to the process undergone by bile ducts in primary biliary cirrhosis and primary sclerosing cholangitis. However, evidence of angiogenesis has been observed in primary biliary cirrhosis and autoimmune hepatitis (Fig. 2). Angiogenesis may occur at a late stage, in response to the hypoxia caused by depletion of existing vessels, although the precise kinetics of the process have not been investigated.
Primary biliary cirrhosis is characterized by chronic patchy destruction and obliteration of septal and intrahepatic bile ductules associated with portal tract infiltration, progressive fibrosis, and intrahepatic cholestasis; the end stage is established cirrhosis.103 An increased number of vascular structures appear in inflamed portal tracts of patients with primary biliary cirrhosis in comparison with controls (Fig. 2 and Moreno-Otero, R., manuscript in preparation). Inducible NO synthase is upregulated in the livers of patients with primary biliary cirrhosis, leading to the local formation of NO, which may contribute to the vasodilation required during the initial phases of angiogenesis.104, 105 VEGF, which is overexpressed in primary biliary cirrhosis tissue samples, may induce NO-mediated vasodilation.14 Further evidence supporting an active angiogenic process in primary biliary cirrhosis is the finding of ECs of small periductular vessels showing positive immunostaining for β1 integrins and a marked staining of ECs for fibronectin and laminin (which directly regulate the angiogenic response).106 Although these findings have been primarily associated with the recruitment of inflammatory cells and the pathogenesis of bile duct destruction, they may also participate in the angiogenic process that characterizes primary biliary cirrhosis.
Autoimmune hepatitis is a chronic progressive liver disease characterized by serological changes (hypergammaglobulinemia, autoantibodies) and interface hepatitis on histological analysis. Tubular structures reflecting formation of new vasculature are observed in inflamed portal tracts of autoimmune hepatitis patients.107 As in primary biliary cirrhosis and chronic hepatitis C, an upregulation of inducible NO synthase leading to an enhanced production of NO occurs in autoimmune hepatitis, and this may participate in the angiogenic process.104 However, information about the mechanisms of angiogenesis in autoimmune hepatitis is still very limited.
Alcoholic Liver Disease.
Hypoxia is one of the pathogenic mechanisms contributing to liver damage secondary to acute and chronic alcohol consumption.108–110 In a rat experimental model of alcoholic liver disease, hypoxia is evidenced as an increase in adduct formation in periportal and centrilobular areas using pimonidazole (a hypoxyprobe), particularly during peaks of blood alcohol levels.111 Local low levels of oxygen have been postulated not only to contribute to liver damage, but also to induce several factors involved in angiogenesis: e.g., VEGF, inducible nitric oxide synthase.111 Acute ethanol intake also induces the formation of large gaps coexisting with normal sieve-plate fenestrations in sinusoidal ECs, suggesting that intercellular contacts are broken, and chronic ethanol feeding reduces the number and disrupts the natural organization of fenestrae.112 These findings reflect the existence of an angiogenic process. However, EC proliferation has been reported to be inhibited during experimental alcoholic liver disease in rats, an effect mediated by increased TGF-β levels.113
Conclusions and Future Perspectives
In recent years, it has become increasingly evident that pathological neoangiogenesis processes occur in chronic inflammatory and fibrotic liver diseases (and probably will be discovered in other nontumoral hepatic alterations). The challenge for the upcoming years is the characterization of the molecular basis and pathways of angiogenic disorders in an integrated manner. If the currently uncertain pathogenetic role of angiogenesis in chronic inflammatory liver diseases can be demonstrated, it could have a prognostic value in the evaluation of disease progression. The identification of reliable and easily measurable molecular markers reflecting the degree of vascularization would then be a major goal for research. A better understanding of the process may also lead to the design of efficient and safe antiangiogenic therapies using appropriate combinations of inhibitors of angiogenesis. The clinical response can be expected to be greater than in tumoral conditions, given that nontransformed cells are less likely to activate alternative angiogenic pathways conferring resistance to therapy. Additionally, modulation of angiogenesis by current antifibrotic, anti-inflammatory, or autoimmune therapies in chronic liver diseases can also be expected.
The authors thank Brenda Ashley for assistance with English and Juan A. Martín (Acción Médica) for help with illustrations.