Hepatocyte growth factor twenty years on: Much more than a growth factor


  • Takahiro Nakamura,

    1. Division of Tumor Dynamics and Regulation, Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa, Japan
    Search for more papers by this author
  • Katsuya Sakai,

    1. Division of Tumor Dynamics and Regulation, Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa, Japan
    Search for more papers by this author
  • Toshikazu Nakamura,

    1. Kringle Pharma Joint Research Division for Regenerative Drug Discovery, Center for Advanced Science and Innovation, Osaka University, Suita, Osaka, Japan
    Search for more papers by this author
  • Kunio Matsumoto

    Corresponding author
    1. Division of Tumor Dynamics and Regulation, Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa, Japan
      Dr. Kunio Matsumoto, Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan. Email: kmatsu@staff.kanazawa-u.ac.jp
    Search for more papers by this author

  • Conflict of interest
    KM has acted as a Cheif Scientific officer and own stocks in Kringle Pharma, Inc.

Dr. Kunio Matsumoto, Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan. Email: kmatsu@staff.kanazawa-u.ac.jp


Liver regeneration depends on the proliferation of mature hepatocytes. In the 1980s, the method for the cultivation of mature hepatocytes provided an opportunity for the discovery of hepatocyte growth factor (HGF) as a protein that is structurally and functionally different from other growth factors. In 1991, the scatter factor, tumor cytotoxic factor, and 3-D epithelial morphogen were identified as HGF, and Met tyrosine kinase was identified as the receptor for HGF. Thus, the connection of apparently unrelated research projects rapidly enriched the research on HGF in different fields. The HGF-Met pathway plays important roles in the embryonic development of the liver and the placenta, in the migration of myogenic precursor cells, and in epithelial morphogenesis. The use of tissue-specific knockout mice demonstrated that in mature tissues the HGF-Met pathway plays a critical role in tissue protection and regeneration, and in providing less susceptibility to chronic inflammation and fibrosis. In various injury and disease models, HGF promotes cell survival, regeneration of tissues, and suppresses and improves chronic inflammation and fibrosis. Drug development using HGF has been challenging, but extensive preclinical studies to address its therapeutic effects have provided significant results sufficient for the development of HGF as a biological drug in the regeneration-based therapy of diseases. Clinical trials using recombinant human HGF protein, or HGF genes, are in progress for the treatment of diseases.

Introduction: Before 1991

A mysterious phenomenon, perhaps even now, is the vital ability of the liver to regenerate. After experimental partial hepatectomy in mice and rats, for which two-thirds of the liver is removed, the residual liver tissue enlarges to make up for the mass of the removed tissue. The entire process can be completed within only one week. An in vitro cell culture technique for rat hepatocytes provided a scientific background to help explain liver regeneration. In 1984, hepatocyte growth factor (HGF) was discovered as a mitogenic protein for rat hepatocytes,1,2 and was thereafter purified from rat platelets,3 human plasma,4 and rabbit plasma.5

In 1989, cDNA for human HGF was cloned and the primary structure of HGF was clarified, by which HGF was identified as a novel growth factor with unique structural characteristics.6,7 Biologically active HGF, a protein composed of 697 or 692 amino acids, is a heterodimeric molecule composed of an α-chain and a β-chain (Fig. 1a). The α-chain contains four kringle domains, while the β-chain contains a serine protease-like structure. A striking fact is that HGF has a structural similarity to plasminogen, which is a heterodimeric serine protease containing five kringle domains.

Figure 1.

Structural characteristic of hepatocyte growth factor (HGF) (a) and Met (b). Sema, the domain found in semaphorin receptors; PSI, the domain found in plexins, semaphorins and integrins; IPT, the domain found in immunoglobulins, plexins and transcription factors.

In 1991, we contributed to this journal by providing our review article entitled “Hepatocyte growth factor: molecular structure and implications for a central role in liver regeneration.”8 In that review we described the structural characteristics of HGF, changes in its expression following liver injury, and newly identified target cell types and the biological action of HGF. Before 1991, studies on HGF were done only by a small number of research groups, and total publications on HGF numbered less than 20 (Table 1). Research on HGF has widely spread and currently over 300 publications appear each year worldwide. After 20 years, recombinant human HGF is currently in clinical trials for treatment of diseases. In this review, we attempt to track 20 years of HGF research and development.

Table 1.  Landmark events in HGF research
YearEventsReferences (ref. No.)
1984Identification and partial characterization of HGF as mitogenic protein for hepatocytesNakamura et al.1
Russell et al.2
1987Purification of rat HGF from plateletsNakamura et al.3
Identification and partial characterization of scatter factor as fibroblast-derived cell motility factorStoker et al.9
1988Purification of human HGF from plasmaGohda et al.4
1989Purification of rabbit HGF from plasmaZarnegar & Michalopoulos5
Nakamura et al.6
cDNA cloning for human HGFMiyazawa et al.7
Purification of mouse scatter factor from culture supernatant of fibroblastsGherardi et al.10
1990Purification of human scatter factor from culture supernatant of fibroblastsWeidner et al.11
1991Establishment of notion that scatter factor is identical molecule to HGFseveral reports
Bottaro et al.12
Identification of c-Met protooncogene product as receptor for HGFNaldini et al.13
Identification of epithelial 3D-morphogen as HGFMontesano et al.14
Identification of tumor cytotoxic factor as HGFShima et al.15
1992In vivo therapeutic actions of HGF using recombinant HGFIshiki et al.16
1995Generation of knockout mice disrupted with HGF or Met geneSchmidt et al.17
Uehara et al.18
Therapeutic action of HGF for liver cirrhosis as fibrotic and chronic inflammatory disorderBladt et al.19
Matsuda et al.20
1997Identification of germline mutations in Met gene in patients with hereditary and sporadic papillary renal cancerSchmidt et al.21
Isolation of NK4 as specific inhibitory molecule for HGF-Met pathwayDate et al.68
1998Determination of crystal structure of NK1 (N-terminal and the first kringle domains)Ultsch et al.22
In vivo therapeutic action of a specific inhibitor for HGF-Met in experimental cancer modelDate et al.23
1999Determination of crystal structure of NK1-dimer structureChirgadze et al.24
2002Clinical study for treatment of patients with chronic leg ulcer with recombinant HGF proteinNayeri et al.25
2003Determination of crystal structure of Met tyrosine kinase domainSchiering et al.26
2004Determination of crystal structure of the HGF β-chain in complex with the Sema-PSI domains of MetStamos et al.27
Clinical study for treatment of patients with critical limb ischemia by HGF gene drugMorishita et al.28
2009Clinical trial for safety evaluation using recombinant HGF protein drughttp://www.kringle-pharma.com/en/
2010Double-blind clinical trial for treatment of patients with critical limb ischemia by HGF gene drugShigematsu et al.29

In 1991

In 1991, there was unexpected landmark progress in the research on HGF (Table 1). The scatter factor, originally identified as a fibroblast-derived cell motility factor for epithelial cells,9 was shown to be an identical molecule to HGF. Similarly, tumor cytotoxic factor, a fibroblast-derived factor that induces cell death for several cancer cell types, was shown to be an identical molecule to HGF.15 These bioactive molecules were identified and characterized using biological assays from different groups.

The induction of branching tubulogenesis in renal epithelial cells by HGF had a particular impact on cell and developmental biologists, because among growth factors and bioactive molecules HGF was the first to induce 3-dimensional (3-D) morphogenesis.14 It had long been recognized that morphogenesis of epithelial tissues required an interaction with the adjacent mesenchyme (during embryonic development) or with the adjacent stroma (during postnatal life), which is known as the epithelial-mesenchymal interaction. However, the molecule that mediated the inducing effect of mesenchyme or stroma on epithelial morphogenesis was yet to be identified. HGF was the hitherto unidentified mesenchymal-derived molecule in epithelial-mesenchymal interaction that was responsible for epithelial morphogenesis.

The receptor for HGF was identified as a product of the c-met proto-oncogene in 1991.12,13 The Met receptor is composed of structural domains that include the extracellular Sema, PSI and IPT domains, the transmembrane domain, and the intracellular juxtamembrane and tyrosine kinase domains (Fig. 1b). The binding of HGF to the Met receptor induces activation of Met tyrosine kinase and the autophosphorylation of tyrosine residues in Met.

Scientific research sometimes unexpectedly turns in unanticipated directions. The connection and integration of research and molecules previously thought to be unrelated, i.e. HGF, scatter factor, tumor cytotoxic factor, 3-D epithelial morphogen, and Met, rapidly enriched the research on HGF in different fields.

3-D Morphogenesis and organogenesis in development

The essential roles of the HGF-Met pathway in mammalian development have been defined by the targeted disruption of the HGF or Met genes.17–19 In these studies, the embryonic liver is reduced in size and shows extensive apoptotic cell death. The knockout is embryonically lethal due to impaired organogenesis of the placenta and liver. In the placenta, the number of labyrinthine trophoblasts is markedly reduced. It is noteworthy that HGF-Met participates in a long-distance migration of cells in development, which indicates a particular role for HGF in cell movement. In the mouse embryo, HGF is strongly expressed in the limb bud mesenchyme and septum transversum (which develops into the diaphragm), and the migration of Met-positive myogenic precursor cells from dermo-myotome in the somite to limb buds and diaphragm is impaired in Met-/- embryos. Consequently, skeletal muscles of the limb and diaphragm are not formed in mutant mice.19 Thus, HGF provides spatially defined chemoattractant-like motogenic signals for the migration of myogenic precursor cells.30

Using a gene knock-out/knock-in approach and sterotaxic injections of HGF or neutralizing antibody into the striatum in mice reveals the critical role of HGF in the development of the nervous systems.31 HGF functions as an axonal chemoattractant for spinal motor neurons and for the projection of motor neurons to limb muscle.32 In addition, HGF plays important roles in the development of sensory, sympathetic, parasympathetic, and cortical neurons. HGF regulates the proliferation of oligodendrocyte progenitor cells and their differentiation into oligodendrocytes.33 Thus, HGF is implicated in neuronal as well as glial development in an orchestrated manner.

HGF regulates epithelial development and morphogenesis in different organs.30,34 In organ culture experiments, antibodies against HGF inhibit branching tubulogenesis of developing epithelia in the kidney and mammary glands. In tooth germ culture, antisense oligonucleotide to HGF induces impaired morphogenesis of tooth epithelium, which subsequently differentiates into ameloblasts. In organ cultures of the developing lung, branching morphogenesis of developing lung epithelia is inhibited by neutralization of HGF or use of an antisense strategy. During the development of various tissues, Met is expressed in the epithelia, while the HGF in mesenchymal cells is in close vicinity in various organs. These expression patterns indicate that HGF is a mesenchymal-derived factor that predominantly acts on neighboring developing epithelia.30 Interactions between epithelium and mesenchyme, i.e. epithelial-mesenchymal interactions mediate crucial aspects of development, affecting tissue induction and epithelial morphogenesis. Thus, HGF plays important roles as a mesenchymal-derived factor that regulates epithelial growth and morphogenesis.

Physiological roles

Far beyond the initial prediction that HGF plays a particular role in the regulation of liver growth and regeneration, the diverse biological and physiological roles of the HGF-Met pathway have been studied in diverse cell types (Table 2). In addition to the mitogenic, motogenic (enhancement of cell motility), and 3-D morphogenic activities of HGF—those known prior to 1991—an increasing number of studies done in the early 1990s on apoptosis were connected to the remarkable action of HGF in the suppression of apoptotic cell death (Fig. 2). Activation of the Fas receptor by agonistic antibody in mice induces fulminant hepatitis associated with massive apoptosis of hepatocytes, whereas HGF potently suppresses apoptotic death of hepatocytes, thereby preventing the onset of fulminant hepatitis in mice.35 The suppression of cell death by HGF participates in the biological, physiological, and therapeutic actions of HGF. The impaired development of the embryonic livers of systemic knockout mice with the HGF or Met was explained by submassive apoptosis in the hepatoblasts. The cytoprotective action of HGF explains how it suppresses the onset of tissue damage, including acute liver injury. However, it is noteworthy that HGF bidirectionally regulates cell survival in a cell type-dependent manner. In a model of liver cirrhosis in rats, HGF suppressed apoptosis of hepatocytes but facilitated apoptosis of α-smooth muscle and actin-positive myofibroblasts, the cells responsible for tissue fibrosis.36

Table 2.  Target cell types of HGF
Tissue typeTarget cell type
Hepato-biliary and pancreasHepatocyte
Bile duct epithelial cell
Pancreatic β cell
GastrointestinalGastric epithelial cell
Intestinal mucosal epithelial cell
KidneyRenal tubular cell
LungBronchial epithelial cell
Alveolar type II epithelial cell
NervousNeuron (hippocampal neuron, cerebral cortex neuron, midbrain dopaminergic neuron, cerebellar granular neuron, motor neuron, thalamic neuron, sensory neuron, sympathetic neuron, parasympathetic neuron, subventiricular zone neural stem-like cell)
Schwann cell
Oligodendrocyte progenitor cell
Cardiovascular and lymphaticCardiomyocyte (in hypoxic condition)
Vascular endothelial cell
Lymphatic vessel cell
Hematopoietic and immuneDendritic cell
Hematopoietic stem/progenitor cell
Macrophage (conditionally)
Skin and eyeKeratinocyte
Hair bulb keratinocyte
Corneal epithelial cell
Muscle, bone, and jointMuscle satellite cell
Myogenic precursor cell
Articular chondrocyte
GlandsMammary gland epithelial cell
Submandibular gland epithelial cell
Salivary gland epithelial cell
Prostate epithelial cell
Thyroid cell
Figure 2.

Outline for biological and physiological actions of hepatocyte growth factor (HGF), which participates in tissue protection and regeneration by HGF, and therapeutic actions of HGF.

Induction of proteases involved in breakdown of the extracellular matrix scaffold was also revealed as a particular action of HGF. HGF induces or up-regulates expression of urokinase-type plasminogen activator and matrix metalloproteinases (MMPs) such as membrane-type MMP and MMP-9. Induction of these proteases also participates in the biological, physiological, and therapeutic actions of HGF (Fig. 2). For 3-D epithelial branching tubulogenesis by HGF, activation of urokinase-type plasminogen activator and membrane type 1-MMP play a crucial role. Enhancement in the expression of MMPs is a mechanism by which the fibrotic change in tissues is suppressed and/or the resolution of fibrosis is facilitated by HGF. Instead, the induction of MMPs, such as membrane types 1-MMP and MMP-9, participates in the 3-D spreading and invasion of cancer cells.

In accordance with the initial implication of HGF as a humoral hepatotrophic factor that enhances liver regeneration, expression of HGF is increased in response to liver injuries. Conversely, neutralization of endogenous HGF enahances liver damage, for example by increasing apoptosis/necrosis of hepatocytes and/or suppression of liver regeneration. The hepatotrophic role of HGF was definitively demonstrated using a conditional knockout of the Met gene in mice.37–39 Mice lacking the Met gene in hepatocytes were hypersensitive, even to mild liver injury caused by administration of a low-dose of agonistic anti-Fas antibody, indicating that anti-apoptotic activity of HGF plays a role in protection of the liver. Liver- or hepatocyte-specific Met-/- mice showed delayed liver regeneration associated with persistent inflammatory reaction, and were susceptible to fibrotic change in the liver. Likewise, after bile duct ligation that causes chronic cholestatic liver injury, hepatocyte-specific Met-/- mice showed increases in hepatocyte apoptosis, inflammation, and profibrogenic responses, and were more susceptible to chronic inflammation and fibrotic change compared with control mice.39 These effects in liver- or hepatocyte-specific Met-/- mice clearly indicate that the physiological roles of the HGF-Met pathway in protection, regeneration, anti-inflammation, and anti-fibrosis of the liver cannot be substituted by other growth factors, cytokines, and bioactive molecules. Thus, the hepatotrophic and hepatoprotective roles of HGF have been well established during the past 20 years.

Similar to the story of HGF-Met in the liver, the involvement of the HGF-Met pathway in tissue protection and/or regeneration has been demonstrated in different tissues, though the dependency on the HGF-Met pathway is different depending on tissue types. The HGF-Met pathway supports the protection and regeneration of kidney, lung, nervous system, cardiovascular, cutaneous, and gastrointestinal tissues. Collectively, the HGF-Met pathway plays definitive roles not only by promoting survival, proliferation, migration, and 3-D morphogenesis but also in preventing inflammation and fibrotic change in tissues. Studies using tissue-specific Met knockout mice provided clear evidence for the roles of the HGF-Met pathway in protection, regeneration, and anti-fibrosis/inflammation in different cell and tissue types.

Mice with conditional knockout of Met in the collecting duct of the kidney were more susceptible to interstitial fibrosis and tubular necrosis after unilateral ureteral obstruction, while they had reduced capacity in tubular cell regeneration after release of the obstruction, leading to diminished functional recovery.40 In podocyte Met conditional knockout mice, no pathology was seen, whereas the mice developed more severe podocyte apoptosis and albuminurea in comparison with control mice.41 Disruption of the Met gene in epidermal keratinocytes demonstrated in an indispensable role for the HGF-Met pathway in skin wound healing.42 Surprisingly, Met-deficient epidermal keratinocytes were unable to contribute to the re-epithelialization of skin wounds, though other growth factors and bioactive molecules were functional.

Conditional knockout mice with selective disruption of Met in pancreatic β-cells displayed significantly reduced plasma insulin after a glucose challenge. In vitro glucose-stimulated insulin secretion in the islets from β-cell-Met-/- mice was decreased by ∼50% compared with control islets. These changes in β-cell function in conditional Met knockout mice were not accompanied by changes in total β-cell mass, islet morphology, and β-cell proliferation.43 Another group using β-cell-Met-/- mice displayed mild hyperglycemia and a complete loss of acute-phase insulin secretion in response to glucose.44 Therefore, HGF-Met signaling in the β-cell is not essential for β-cell growth, but it is essential for normal glucose-dependent insulin secretion and glucose homeostasis.

Because aberrant activation of growth factor receptors is generally associated with tumor development, a lack of growth factor receptor-mediated signal could be associated with less tumor development. However, a role of the HGF-Met pathway in hepatocarcinogenesis has been debated. Unexpectedly, when compared with control mice, liver-specific Met-/- mice treated with N-nitrosodiethylamine developed significantly more tumors that were larger and had a shorter latency. N-nitrosodiethylamine induced oxidative stress, whereas administration of antioxidant blocked the hepatocarcinogenesis in liver-specific Met-/- mice.45 Thus HGF-Met signaling is essential for maintaining normal redox homeostasis in the liver and has tumor suppressor effect(s) in this model. In a model of hepatocarcinogenesis induced by N-nitrosodiethylamine and phenobarbital, liver-specific Met-/- mice showed a higher prevalence of macroscopically visible liver tumors, while there were only minor differences in the number of preneoplastic and neoplastic lesions in response to phenobarbital-induced promotion. These results indicate that a defect in Met-mediated signaling increases chemically induced tumor initiation in liver but does not significantly affect phenobarbital-mediated tumor promotion.46

Processing and physiological relevance

HGF is biosynthesized as a prepro-form of 728 amino acids, including a signal sequence and both α- and β-chains. After cleavage of a signal peptide of the first 31 amino acids, a single-chain HGF is further cleaved between Arg494 and Val495, and this processing is coupled to the conversion of biologically inactive pro-HGF to active HGF. It has been proposed that several proteases in the serum or cell membranes are involved in the activation of single-chain HGF, including HGF activator (HGF-A), urokinase-type plasminogen activator, plasma kallikrein, coagulation factors XII and XI, matriptase, and hepsin. Among them, HGF-A, matriptase, and hepsin are the most efficient in processing proHGF. HGF-A was purified and cloned as a serum-derived protease that activates HGF.47,48 HGF-A is a member of the kringle-containing serine protease superfamily. HGF-A is biosynthesized primarily by hepatocytes and circulates in the plasma as an inactive single-chain proHGF-A. Because significant activation of proHGF occurs in injured tissues and thrombin activates proHGF-A, the conversion of prothrombin to thrombin, which occurs during activation of the coagulation cascade, is involved in the activation of HGF in response to tissue injury. In addition, kallikrein 1-related peptidases participate in the activation of proHGF-A, and this activation is believed to occur in the pericellular microenvironment.

The activity of HGF-A is regulated not only by its processing from proHGF-A, but also by a specific inhibitor, HGF-A inhibitor-1 (HAI-1).49 HAI-1 is a membrane-bound serine protease inhibitor that is mainly expressed on the basolateral surfaces of epithelial cells. HGF-A can be localized to the pericellular microenvironment via its affinity to heparansulfate and/or binding to HAI-1. The activity of HGF-A is suppressed by its binding to the cell surface HAI-1. On the other hand, the HGF-A and HAI-1 complexes on the cell surface can potentially be released by metalloprotease-mediated shedding of the HAI-1 ectodomain, and this process is enhanced by inflammatory cytokines such as interleukin-1β. Perhaps, HAI-1 plays a role not only as an inhibitor but also as a cell-associated reservoir of HGF-A. The regulatory mechanism for the activation of proHGF by HGF-A and HAI-1 can be considerable as the physiological link between inflammation and tissue regeneration. In this context, it is notable that inflammatory mediators such as interleukin-1β, prostaglandin E2, and prostaglandin I2 are potent inducers for the expression of HGF, again implicating a physiological link from inflammation to tissue regeneration.

The physiological significance of HGF-A has been studied by loss-of-function approaches.48,49 The inhibition of HGF-A by neutralizing antibody resulted in impairment in activation of proHGF.48 The knockout approach using mice lacking HGF-A indicated that the sera from HGF-A-/- mice were unable to activate proHGF, indicating that HGF-A is the major protease responsible for the activation of HGF in serum.50 Although HGF-A-/- mice showed normal development, the attenuation of initial regeneration after mucosal injury was associated with impaired restitution of epithelia. Pathophysiological study has indicated that fibroblasts from patients with idiopathic pulmonary fibrosis have a lower capacity to activate pro-HGF compared with control fibroblasts.51 Therefore, the process for activation of HGF also plays an important role in the regulation of tissue regeneration and susceptibility to pathological conditions.

Therapeutic approaches and clinical development

The highlights in research on HGF during the past 20 years have resulted in extensive therapeutic approaches using different disease models for different tissues (Table 3). Even in different injury and tissue types, the mechanisms responsible for the therapeutic effects of HGF are likely to overlap (Fig. 2). The protective actions of HGF are explained by prevention of cell death against various types of stresses and injury. Prevention of cell death by HGF seems to be associated with less subsequent inflammation, or an anti-inflammatory effect, whereas mechanisms by which HGF could regulate immunological responses have yet to be addressed. Recent studies have indicated that HGF regulates the function of immune cells such as dendritic cells and a subset of regulatory T cells.52–54 These biological actions of HGF on immune cells are likely to be the underlying mechanism, at least in part, by which HGF exerts therapeutic effects on diseases associated with allergy, inflammation, and fibrosis.

Table 3.  Therapeutic approaches with recombinant HGF in various disease models in various tissues
Tissues and disease modelsModel for injury or diseasesObserved therapeutic effectsReferences
  1. ALT, alanine aminotransferase; AST, aspartate aminotransferase; TNF, tumor necrosis factor.

 Acute hepatitisα-Naphthylisothiocyanate or CCl4Enhancement of hepatocyte proliferationIshiki et al., Hepatology, 1992; 16: 1227
Suppression of serum ALT
α-NaphthylisothiocyanateDecreases in serum ALT, AST, bilirubinRoos et al., Endocrinology, 1992; 131: 2540
Suppression of parenchymal lesions
DimethylnitrosamineSuppression of serum ALT, ASTMasunaga et al., Eur J Pharmacol, 1998; 342: 267
Ischemia-reperfusionSuppression of mortality, serum ALT, and necrotic hepatocytesSakakura et al., J Surg Res, 2000; 92: 261
Inhibition of neutrophil infiltration
Enhancement of hepatocyte growth
Ischemia-reperfusionSuppression of necrotic hepatocytes, serum ALT, ASTOe et al., J Hepatol, 2001; 34: 832
Suppression of oxidative stress
α-Naphthylisothiocyanate + partial hepatectomyEnhancement hepatocyte proliferationYoshikawa et al., J Surg Res, 1998; 78: 54
Decrease in serum bilirubin
 CholestasisBile duct ligationReduction in necrotic and apoptotic hepatocytesLi et al., Am J Physiol, 2007; 292: G639
Decrease in serum AST, ALT
 Fulminant hepatitisAgonistic anti-FAS antibodySuppression of hepatocyte apoptosis, serum ALT, and mortalityKosai et al., Biochem Biophys Res Commun, 1998; 244: 6382
LPS + D-galactosamineSuppression of hepatocyte apoptosis, serum ALT, and mortalityKosai et al., Hepatol, 1999; 30: 151
 Liver cirrhosisDimethylnitrosamineImprovement of survival and liver functionMatsuda et al., J Biochem, 1995; 118: 643
Reduction in extracellular matrix accumulation
Decrease in serum AST, ALT
CCl4, dimethylnitrosamine, or porcine serumImprovement of survival and liver functionMatsuda et al., Hepatol, 1997; 26: 81
Reduction in extracellular matrix accumulation and serum AST level
ThioacetamideReduction in extracellular matrix accumulationOe et al., J Control Release, 2003; 88: 193
DimethylnitrosamineReduction in extracellular matrix accumulation and myofibroblastsKim et al., Am J Pathol, 2005; 166: 1017
Increase in apoptosis in myofibroblasts
DimethylnitrosamineReduction in extracellular matrix accumulationKusumoto et al., Int J Mol Med, 2006; 17: 503
Decrease in serum ALT and TGF-β levels
Increase in serum albumin and liver weight
 Liver cirrhosis + hepatic surgeryDimethylnitrosamine + portal branch ligation + partial hepatectomyPromotion of survivalKaido et al., Hepatol, 1998; 28: 756
Decrease in serum AST, ALT, bilirubin
Increase in liver weight
 Alcoholic steatohepatitisEthanol-containing dietDecrease in hepatic lipidsTahara et al., J Clin Invest, 1999; 103: 313
Increase in serum lipids and lipoproteins
 Ulcerative colitisDextran sulfate sodiumSuppression of histological damage and loss of weightTahara et al., J Pharmacol Exp Therapeutics, 2003; 307: 146
Enhancement of epithelial cell proliferation
2,4,6-trinitrobenzene sulfonic acidSuppression of colonic ulcer coverage and large intestinal shorteningNumata et al., Inflamm Bowel Dis, 2005; 11: 551
Reduction in inflammatory cells and enhancement of epithelial cell growth
 Gastric ulcerCryoinjuryEnhancement of epithelial cell proliferationSchmassmann et al., Gastroenterol, 1997; 113: 1858
 Gastric injuryCisplatin Nakahira et al., Biochem Biophys Res Commun, 2006; 341: 897
 Acute kidney injuryHgCl2 or cisplatinSuppression of renal injuryKawaida et al., Proc Natl Acad Sci USA, 1994; 91: 4357
Enhancement of tubular proliferation
IschemiaEnhancement of tubular proliferationMiller et a., Am J Physiol, 1994; 266: F129
Cyclosporin AEnhancement of tubular proliferationAmaike et al., Cytokine, 1996; 8: 387
Suppression of tubular pathology (vacuolization)
HgCl2Suppression of renal dysfunctionYamasaki et al., Nephron, 2002; 90: 195
Suppression of tubular apoptosis
GlycerolSuppression of tubular necrosis and improvement of renal functionNagano et al., Nephron, 2002; 91: 730
Prevention of mortality
Tacrolims/FK506Suppression of injury and decrease in serum creatinineTakada et al., Transpl Int, 1999; 12: 27
Enhancement of renal cell proliferation
 Acute renal inflammationTumor necrosis factor-αDecrease in sequestration of circulating macrophages in the kidneyGong et al., Kidney Int, 2006; 69: 1166
Suppression of acute renal inflammation
 Septic acute renal failureLipopolysaccharideSuppression of mortalityKamimoto et al., Biochem Biophys Res Commun, 2009; 380: 333
Suppression in blood urea nitrogen and AST
 Diabetic nephropathyStreptozotocinDecrease in albuminureaMizuno & Nakamura, Am J Physiol, 2004; 286: F134
Suppression of glomerular and tubulointerstitial fibrosis
Improvement of renal function
 Chronic kidney diseaseSpontaneous due to tensin2 mutationImprovement of glomerular and tubulointerstitial fibrosisMizuno et al., J Clin Invest, 1998; 101: 1827
Enhancement of tubular proliferation
Improvement of renal function and decrease in albuminurea
5/6 nephrectomySuppression of renal inflammationGong et al., J Am Soc Nephrol, 2006; 17: 2464
Decrease in sequestration of circulating macrophages in the kidney
Unilateral ureteral obstructionImprovement of tubulointerstitial fibrosisMizuno et al., Kid Int, 2001; 59: 1304
Increase in tubular proliferation and decrease in tubular apoptosis
Unilateral ureteral obstructionImprovement of tubulointerstitial fibrosisYang & Liu, Am J Physiol, 2003; 284: F349
 GlomerulonephritisAnti-Thy 1.1 antibodySuppression of mesangial cell proliferationBessho et al., 2003; Am J Physiol, 284: F1171
 Chronic allograft nephropathyIschemia and transplantationPrevention of tubular cell death after ischemia and transplantaionAzuma et al., J Am Soc Nephrol, 2001; 12: 1280
Suppression of proteinurea and fibrotic change of the kidney
Prevention of mortality
 Critical limb ischemiaHindimb ischemiaEnhancement of collateral blood vessel formationVan Belle et al., Circulation, 1998; 97: 381
Increase in blood vessel density and blood flow
Hindimb ischemiaEnhancement of collateral blood vessel formationMorishita et al., Hypertension, 1999; 33: 1379
Hindimb ischemiaImprovement in the recovery of blood flow by slow release deliveryMarui et al., J Vasc Surg, 2005; 41: 82
 Neointimal hyperplasiaBalloon injuryReduction in intimal areaYasuda et al., Circulation, 2000; 101: 2546
Enhancement of regeneration of endothelial cell layer
 Coronary artery diseaseChronic ischemiaImprovement in regional myocardial blood flowYamaguchi et al., Surg Today, 2005; 35: 855
Improvement in myocardial function
 Myocardial infarctionIschemia-reperfusionReduction in infract area, apoptosis in cardiomyocytes, and mortalityNakamura et al., J Clin Invest, 2000; 106: 1511
Improvement of cardiac function
Ischemia-reperfusionPromotion of improvement in cardiac functionJin et al., J Pharmacol Exp Therapeutics, 2003; 304: 654
Suppression of myocardial apoptosis and hypertrophy
 Cardiac allograft vasculopathyIschemia and transplantationPromotion of survival of allograftsYamaura et al., Circulation, 2004; 110: 1650
Suppression of myocardial inflammation apoptosis
Prevention of cardiac allograft vasculopathy and interstitial fibrosis
 Dilated cardiomyopathyMutation in δ-sarcoglycanImprovement of cardiac fibrosis and echocardiographic functionNakamura et al., Am J Physiol, 2005; 288: H2131
Suppression of myocardial apoptosis and hypertrophy
 Acute lung injuryIntratracheal HCl infusionEnhancement of proliferation of airway and alveolar epithelial cellsOhmichi et al., Am J Physiol, 1996; 270: L1031
Ischemia-reperfusionSuppression of histological damageMakiuchi et al., J Heart Lung Transplant, 2007; 26: 935
Decrease in apoptosis
 Lung fibrosisBleomycinSuppression and improvement of fibrosisYaekashiwa et al., Am J Respir Crit Care Med, 1997; 156: 1937
BleomycinDecrease in extracellular matrix depositionDohi et al., Am J Respir Crit Care Med, 2000; 162: 2302
Enhancement in proliferation of epithelial cells
BleomycinSuppression and improvement of fibrosisMizuno et al., FASEB J, 2005; 19: 580
 Pulmonary emphysemaElastaseRegeneration of alveolar structureIshizawa et al., Biochem Biophys Res Commun, 2004; 324: 276
Promotion of recruitment of bone marrow-derived progenitor cells into alveolar epithelial and endothelial cells
 ResectionLeft peumonectomyEnhancement of alveolar and airway epithelial cellsSakamaki et al., Am J Resp Cell Mol Biol, 2002; 26: 525
 Allergic airway inflammation / asthemaSensitization and challenge with ovalbuminSuppression of airway inflammation, collagen deposition, smooth muscle hyperplasia, and remodellingIto et al., Am J Respir Cell Mol Biol, 2005; 32: 268
Reduction in Th2 cytokines and fibrogenic growth factors
Sensitization and challenge with ovalbuminSuppression of eosinophylic airway inflammationOkunishi et al., Int Arch Allergy Immunol, 2009; 149 Suppl. 1: 14
Suppression of antigen-induced allergic immune responses
 Vocal fold scarringRemoval of the lamina propriaBetter vibration interms of mucosal wave amplitude and glottal closureOhno et al., Ann Otol Rhinol Laryngol, 2007; 116: 762
Reduction in collagen deposition and restoration of hyaluronic acid and elastin
Removal of the lamina propriaBetter function (vibration) of laryngeKishimoto et al., Laryngoscope, 2010; 120: 108
Reduction in collagen deposition
 Wound healingFull-thickness cutaneous excision in diabetic micePromotion of wound closure, re-epithelialization, angiogenesis, granulation tissue formationYoshida et al., Growth Factors, 2004; 22: 111
 Promotion of recruitment of neutrophils, monocytes, macrophages, endothelial cells, and re-epithelialization, granulation tissue formation and angiogenesisBevan et al., J Pathol, 2004; 203: 831
Nervous and sensory   
 Cerebral ischemiaOcclusion of carotid arteriesDecrease in delayed neuronal death in the hippocanpusMiyazawa et al., J Cereb Blood Flow Metab, 1998; 18: 345
Transient occlusion of arteriesDecrease in the infarct sizeTsuzuki et al., Neurol Res, 2001; 23: 417
Increase in blood vessel density
Microsphere embolismPrevention of learning and memory dysfunctionDate et al., J Neurosci Res, 2004; 78: 442
Suppression of endothelial cell apoptosis and necrotic tissue damage
Transient forebrain ischemia (the four-vessel occlusion)Suppression of neuronal cell death in hippocanpal neuronsNiimura et al., Neurosci Lett, 2006; 407: 136
Transient middle cerebral artery occlusionDecrease in the infarct sizeShang et al., J Neurosci Res, 2010; 88: 2197
Suppression of apoptotic neurons and increase in autophagic response
 Peripheral nerve injuryHypoglossal nerve axotomySuppression of the loss in choline acetyltransferaseOkura et al., Eur J Neurosci, 1999; 11: 4139
 Amyotrophic lateral sclerosisMutation in superoxide dismutaseSuppression of degeneration of motor neurons and disease progressionIshigaki et al., J Neuropathol Exp Neurol, 2007; 66: 1037
Promotion of survival
 HydrocephalusTransforming growth factor-βSuppression of fibrosisTada et al., Neurobiol Dis, 2006; 21: 576–586
 Retinal injuryIschemia-reperfusionDecrease in apoptosis in ganglion cell layer and inner nuclear layerShibuki et al., Invest Ophthalmol Vis Sci, 2002; 43: 528
Increase in the inner retinal thickness and neuronal function
 Photoreceotr degeneration / Retinitis pigmentosaSodium iodateBetter structural preservation of the outer retinaOhtaka et al., Current Eye Res, 2006; 31: 347
Better functional preservation of retinal pigment epithelium and photoreceptor cells
Phototoxicity (strong fluorescent light)Better the morphological and functional preservation of photoreceptor cellsMachida et al., Invest Ophthalmol Vis Sci, 2004; 45: 4174
Suppression of apoptosis in photoreceptor cells
 Difficulty in hearingNoise-induced hearing lossProtective in the auditory functionReduction in the loss of outer hair cell lossInaoka et al., Acta Oto-Laryngologica, 2009; 129: 453
 Articular cartilage injuryOsteochondral defectsPromotion of repair of osteochondral defectsWakitani et al., Acta Orthop Scand, 1997; 68: 474–480
 Skeletal muscle injuryFreeze damageInhibition of muscle differentiation and retardation of muscle regenerationMiller et al., Am J Physiol, 2000; 278: C174
 Rheumatoid arthritisType II Collagen-induced arthritisEnhancement of Th2-type immune responseOkunishi et al., J Immunol, 2007, 179: 5504
Inhibition of development of collagen-induced arthritis
 Ligament injuryTendon graft into bonePromotion of histological and biomechanical regeneration and improvementNakase et al., Arthroscopy, 2010, 26: 84

On the other hand, promotion of cell proliferation, migration, and 3-D morphogenesis by HGF explains the recovery and re-organization of tissues from injury, whereas the dynamic reconstruction of 3-D tissue structure substantially supports functional recovery. Chronic tissue injury is tightly associated with the onset of fibrotic change, and this is particularly relevant to the pathogenesis of liver cirrhosis and chronic kidney disease; there has been no effective therapeutic approach for the treatment of such chronic fibrotic diseases. It should be emphasized that, at least in animal models, HGF-treatment is highly effective for the treatment of chronic fibrosis in various disease models, including liver cirrhosis, chronic kidney disease, dilated cardiomyopathy, lung fibrosis, and vocal fold scarring.

The first clinical study using recombinant human HGF protein was done to investigate the physiological and therapeutic effects of HGF on chronic leg ulcers. HGF in gel form was locally applied to chronic leg ulcers in 11 patients.25 The first clinical study of HGF gene therapy by naked expression plasmid was done to investigate its safety for treatment of patients with arteriosclerosis obliterans or Buerger disease.28 Subsequently, a multicenter, randomized, double-blind, placebo-controlled clinical trial was performed for the treatment of patients with critical limb ischemia to evaluate the efficacy and safety of HGF gene therapy using naked plasmid.29 This HGF gene therapy was proven safe and effective for critical limb ischemia. Phase-II and Phase-III clinical trials of HGF gene therapy for the treatment of peripheral arterial disease has been completed in both the USA and Japan. The phase-I clinical trial of the systemic administration of recombinant HGF protein and the Phase-I/II clinical trial for the local application of recombinant HGF protein are both in progress (http://www.kringle-pharma.com/en/index.html).

Structural understanding

The C-terminal multifunctional docking site of Met plays a crucial role in the activation of Met-dependent intracellular signal transduction and biological activities.55 The phosphorylation of C-terminal tyrosine residues in the docking site recruit intracellular signaling molecules, including PI3K (phosphatidylinositol 3-kinase), Grb2 (growth-factor-receptor-bound protein 2), Gab1 (Grb2)-associated binder 1), PLCγ (phospholipase Cγ), and Shp2 (SH2-domain-containing protein tyrosine phosphatase 2). Among signaling molecules, a scaffolding adaptor protein, Gab1, is the most crucial substrate for the HGF-Met pathway.56 Direct interaction of Gab1 with tyrosine phosphorylated Met is mediated by the Met-binding site in Gab1, and it allows a direct and robust interaction between Met and Gab1. Knockout mice with the Gab1 gene exhibited phenotypes similar to those seen in HGF and Met knockout mice.56

In terms of Met-dependent signal transduction, the cytoplasmic juxtamembrane domain, which is composed of 47 highly conserved amino acids, acts as a negative regulator. Cbl, an E3 ubiquitin ligase, binds phosphorylated Y1003 of Met, and this Cbl binding results in Met ubiquitination, endocytosis, transport to the endosomal compartment, then degradation.57 Cbl-mediated degradation of the activated Met provides a mechanism that attenuates or terminates Met-mediated signaling. Phosphorylation of Ser985 in the juxtamembrane domain regulates the activation status of Met upon HGF stimulation. Ser985 is phosphorylated by protein kinase-C and is dephosphorylated by protein phosphatase-2A. In cells in which Ser985 is phosphorylated by treatment with protein kinase-C, HGF-induced activation of Met is suppressed.58,59 Therefore, activation of protein kinase-C, which occurs by different types of extracellular stimuli, regulates HGF-dependent Met activation.

HGF binds to Met through two different mechanisms: the α-chain binds with high affinity while the β-chain binds with very low affinity. Among the α-chain binding sites, NK1 (the N-terminal and first kringle domains) in the α-chain of HGF provides high-affinity binding to Met (Fig. 1a). The α-chain alone exhibits high-affinity binding to Met, whereas the binding of the α-chain does not activate Met.60 When Met is occupied by the α-chain, the low-affinity binding of the β-chain induces activation of Met and biological responses. Hence, the α-chain is a high-affinity binding module to Met, while the β-chain is an activation module for Met.

The structure of the complex of HGF β-chain and Sema was revealed by crystallographic analysis (Fig. 3a).27 The Sema domain of Met forms a seven-bladed β-propeller, which makes the shape of the Sema domain resemble a funnel. Generally, in β-propellers each of the blades is formed by four antiparallel β-strands. The HGF β-chain binds to the bottom face of the propeller, and forms contacts with residues that protrude from blades 2 and 3. The HGF β-chain binds to a series of protruding polar side chains from Met, which originate from three separate loops: residues 124–128, residues 190–192, and residues 218–223 (Fig. 3a). Although the α-chain of HGF binds to Met with much higher affinity than that of the HGF β-chain, the crystal structure for the interaction between the HGF α-chain and the extracellular region of Met has yet to be determined.

Figure 3.

Crystal structures for the complex of hepatocyte growth factor (HGF) β-chain and the Met Sema domain (a) and the Met tyrosine kinase domain (b). The crystal structures for the complex of HGF β-chain and the Met Sema domain were reported by Stamos et al. (2003) (PDB number: 1SHY). The crystal structure for Met tyrosine kinase was reported by Schiering et al. (2003) (PDB number 1ROP). In (b), the activation loop (A-loop) is shown in yellow, K-252a in green, and selected tyrosine residues (Y1234F, Y1235D, Y1349, Y1356) are in blue.

The Met tyrosine kinase domain follows the bilobal protein kinase architecture mainly with an N-terminal, β-sheet-containing domain linked through a hinge segment mainly to the α-helical C lobe (Fig. 3b).26,61 The characteristic feature of Met is the presence of the C-terminal multifunctional docking site that contains tyrosine residues (1349YVHVNAT1356YVNV). In an unphosphorylated Met kinase domain, Met 1229 in the activation loop (A-loop, yellow in Fig. 3b) projects into the ATP-binding pocket, and the direction of the Glu 1127 residue involved in ATP-binding is changed, by which ATP is unable to form an appropriate structure for the kinase-substrate complex.61 This structure corresponds to the quiescent autoinhibited Met kinase without HGF stimulation. On the other hand, the autoinhibited Met kinase structure changes upon the phosphorylation of Tyr 1234 and Tyr 1235 in the activation loop, which allows a complex formation between the Met kinase and ATP.

The staurosporine analog K-252a inhibits Met tyrosine kinase through its binding in the ATP pocket.26 Because the structure of the Met kinase domain, complexed with the K-252a as shown in Fig. 3b, was obtained using the recombinant Met kinase domain where Tyr 1234 and Tyr 1235 were replaced by Phe and Asp, respectively, Met kinase complexed with K-252a is expected to represent a structure of Met tyrosine kinase that is activated upon HGF-stimulation.

HGF in tumor biology and therapeutics

The HGF-Met system drives the breakdown of the extracellular matrix and the concomitant cellular migration, mitogenesis, and morphogenesis, leading to the construction of tissue architecture. In cancer tissues, biological programs regulated by the HGF-Met pathway are adopted particularly for invasion and metastasis, which are the life-threatening events of cancer. Early studies indicating the identity of a fibroblast-derived scatter factor in HGF implicated a role for HGF in cancer spreading and invasion.11 The crucial role of stromal fibroblasts in invasion of cancer cells through the 3-D collagen was first demonstrated independently using human oral squamous cell carcinoma.62 The cancer cells invaded aggressively only when they were co-cultured with stromal fibroblasts in collagen gel. The fibroblast-derived factor responsible for the 3-D invasion was identified as HGF. The profound action of HGF on cancer invasion has been demonstrated in a variety of cancer cells, and it has been established that HGF is a mediator in the tumor-stromal interaction that affects the malignant behavior of cancer.63

In addition to the paracrine activation of Met through tumor-stromal interaction, Met activation in cancer often occurs through an autocrine mechanism, or through a mutation in the Met gene. Genetic analysis indicated that missense mutations in the c-met gene are the causative genetic disorder in inherited, and some sporadic papillary, renal carcinomas.21 Mutations found in papillary renal carcinomas are located in the tyrosine kinase domain of the c-Met receptor, and these Met receptor mutations are likely to be a gain-of-function mutation.64 In addition to papillary renal carcinoma, missense mutations in the Met have been found in lung cancer, hepatocellular carcinoma and gastric cancer; sites include the Sema, IPT, juxtamembrane, and tyrosine kinase domains.65

The HGF-Met pathway has become a hot target in anticancer drug development.66,67 Several distinct lines of approach to the inhibition of the HGF-Met pathway have been demonstrated, including small molecules that inhibit Met tyrosine kinase activity, ribozymes, small-interfering RNA (siRNA), neutralizing monoclonal antibodies, soluble Met receptors, and antagonists composed of selected domains in HGF (Fig. 4a). Among these approaches to the inhibition of the HGF-Met pathway, NK4 was the first-identified specific inhibitor for HGF-Met.68,69 NK4 is composed of the N-terminal (N) and four kringle domains (K4) of HGF, and it is a competitive inhibitor of HGF-dependent Met activation (Fig. 4b). Discovery of NK4 as a competitive inhibitor for HGF-Met was soon followed by the experimental treatment of cancer in mice: NK4 inhibited the invasion and growth of gallbladder cancer.23 It is also noteworthy that NK4 is bifunctional—it is an angiogenesis inhibitor as well as an HGF-Met inhibitor.69,70 The inhibition of tumor growth by NK4 treatment has been observed in a variety of tumors, and this inhibitory effect has been associated with a reduction in blood vessels in tumor tissues. NK4 treatment inhibited invasion and metastasis in different types of cancer models, including breast, colon, gastric, lung, ovarian, and pancreatic carcinomas, and malignant melanoma.69

Figure 4.

Outline for distinct approaches targeting HGF and Met for potential cancer treatment (a), outline for biological activity of NK4 as a competitive hepatocyte growth factor (HGF)-antagonist (b), and anti-cancer action of NK4 (c, d). Inhibitory effect of NK4 on tumor invasion (c) and cancer metastasis (d). Photographs in c show invasion of human malignant mesothelioma cells in a 3-D collagen gel. Photographs in d show appearance of mesentery with disseminative metastasis of pancreatic cancer in mice. NK4 inhibited 3-D tumor invasion (c) and cancer metastasis (d).

If metastatic tumors could be suppressed to a non-metastatic state, there would be a considerable improvement in rate of cancer cures. Recent studies have indicated that cancer stem cells participate in drug-resistance and in the spreading of tumors, indicating that cancer stem cells are therapeutic targets for future advancements in anti-cancer strategy.71,72 The activation of the HGF-Met pathway participates not only in the spreading and invasion of cancer stem cells but also in drug resistance to tyrosine kinase inhibitors (e.g. gefitinib and erlotinib) in patients with lung cancer.72–74 Taken together, it seems likely that HGF-Met inhibitors will provide further advances in the molecular targeting therapy of cancer. Several inhibitors of the HGF-Met pathway are under preclinical and clinical development.66,67

Perspectives and conclusions

Current attention and progress in stem cell biology has facilitated an understanding of how stemness of stem cells is controlled by genetic programming, and how stem cells participate in tissue regeneration. A small population of hepatic progenitor cells, or stem cells with self-renewal and slow recycling characteristics, participate in long-term liver growth and the renewal of hepatocytes. However, liver regeneration mainly depends on the proliferation of mature hepatocytes. The key to understanding liver regeneration remains to find out how liver tissue architecture and homeostasis of liver mass are precisely regulated, before, during and after the regenerative response.

Growth factors and their receptor tyrosine kinases are divided into families based on structural and functional similarities. Among each family, a single growth factor activates multiple receptors with structural similarities, while a single growth factor receptor has multiple ligands with structural and functional similarities. By contrast, the sole receptor of HGF is Met, while the sole ligand of Met is HGF; the relationship between HGF and Met is a “one-to-one relationship.” Moreover, as demonstrated in systemic and conditional knockout mice with Met, HGF has unique physiological and therapeutic actions that are not substituted by other growth factors, at least not fully. Taken together, HGF and Met have remarkable potential value as targets in drug discovery and development, from aspects that are both agonistic/activating and antagonistic/inhibiting. The one-to-one relationship for HGF-Met should facilitate future drug discovery and design based on the crystal structures of HGF and/or Met for a small molecular activator or inhibitor of HGF-Met. Perhaps small molecular inducers or suppressors for HGF are also potential candidates in drug discovery.

It has been 20 years since we published our first HGF review article (“Hepatocyte growth factor: molecular structure and implications for a central role in liver regeneration”) in the Journal of Gastroenterology and Hepatology. The progress and diversity of research on HGF during the past 20 years have been remarkable. It is no longer possible to comprehensively review this progress in a single article. Our intent was to describe the structural and functional characteristics and the therapeutic significance of HGF. Drug development using growth factors remains a challenge, but extensive studies to address the therapeutic effects of HGF have provided enough significance to develop agonists and antagonists of the HGF/Met system as biological drugs in regeneration-based or anti-cancer therapies. We propose that the further development of HGF as a therapeutic drug target is well worth the challenge.


The studies from the authors’ laboratories were supported by Grants from the Ministry of Education, Culture, Science, Sports, and Technology of Japan, and by The Program for Promotion of Fundamental Studies in Health Sciences of the National Institute of Biomedical Innovation.