Requirement of Stat3 signaling for HGF/SF-Met mediated tumorigenesis


Hepatocyte Growth Factor/Scatter Factor (HGF/SF) mediates a wide variety of cellular responses by acting through the Met tyrosine kinase receptor. Inappropriate expression of HGF/SF and/or Met has been found in most types of solid tumors and is often associated with poor prognosis. Importantly, constitutional and sporadic activating mutations in Met have been discovered in human papillary renal carcinomas and other cancers, while autocrine and paracrine signaling of this receptor/ligand pair has been shown to contribute to tumorigenesis and metastasis. Numerous downstream signaling molecules have been implicated in HGF/SF-Met mediated tumorigenesis and metastasis. Stat3 is a downstream signaling molecule activated by HGF/SF-Met signaling, and is reported to contribute to cell transformation induced by a diverse set of oncoproteins. Stat3 is constitutively activated in many primary tumors and tumor cell lines, suggesting that signaling by this molecule may be important for cell transformation. To address whether Stat3 is required for HGF/SF-Met mediated tumorigenesis and metastasis, we introduced a dominant-negative form of Stat3, Stat3β into the human leiomyosarcoma cell line SK-LMS-1. We found that Stat3β has no effect on the transformed morphology, proliferation, invasion or branching morphogenesis in vitro. By contrast, expression of Stat3β affected HGF/SF-Met mediated anchorage-independent colony formation and prevented tumorigenic growth in athymic nu/nu mice. Thus, Met signaling through Stat3 provides an essential function for tumorigenic growth, which is manifested in vitro by loss of anchorage-independent growth.


Hepatocyte growth factor/Scatter Factor (HGF/SF) is a multipotent growth factor which induces a wide variety of cell events, including mitogenesis, motility and morphogenesis (Gherardi and Stoker, 1991). All HGF/SF-dependent cell events are mediated by the high affinity membrane tyrosine kinase receptor Met (Bottaro et al., 1991; Weidner et al., 1993). HGF/SF-Met signaling usually acts in a paracrine manner, as HGF/SF is produced by mesenchymal cells, while Met is predominantly expressed in the cells derived from epithelial or endothelial origin (Sonnenberg et al., 1993; Stoker et al., 1987).

Upon HGF/SF binding, Met is tyrosine-phosphorylated and in turn activates its downstream pathways through multiple docking sites within its cytoplasmic domain (Ponzetto et al., 1994). A number of cytoplasmic signaling molecules or adaptor proteins, such as PI3 kinase (Graziani et al., 1991), Grb2 (Ponzetto et al., 1994), Gab1 (Weidner et al., 1996), Shc (Pelicci et al., 1995), PLC-γ (Ponzetto et al., 1994), Shp2 phophatase (Fixman et al., 1996), Src (Faletto et al., 1993; Ponzetto et al., 1994; Rahimi et al., 1998), Akt (Fan et al., 2000; Xiao et al., 2001) and Stat3 (Boccaccio et al., 1998), have been reported mediating HGF/SF-Met dependent signal transductions. HGF/SF-Met signaling is essential for embryogenesis, as targeted disruption in mice of either the gene for HGF/SF ligand or the Met receptor leads to embryonic lethality with defects in liver, placenta, skeletal muscle as well as sensory neuron development (Bladt et al., 1995; Maina et al., 1997; Schmidt et al., 1995; Uehara et al., 1995). HGF/SF also promotes wound healing (Nusrat et al., 1994), organ regeneration (Matsumoto and Nakamura, 1993), neural induction (Streit et al., 1995) and angiogenesis (Bussolino et al., 1992; Grant et al., 1993).

While essential for normal development processes, HGF/SF-Met signaling is frequently implicated in a variety of human cancers. Inappropriate expression of HGF/SF and/or Met has been found in most types of solid tumors including sarcomas, melanomas, carcinomas and gliomas (Cortner et al., 1995; Giordano et al., 1988; Koochekpour et al., 1997; Vande Woude et al., 1997). Establishment of autocrine HGF/SF-Met signaling in NIH3T3 cells leads to cell transformation and induces tumorigenic and metastatic behavior in nude mice (Rong et al., 1992, 1994). Mutations in c-Met gene have been discovered in human papillary renal cell carcinomas (Schmidt et al., 1997), gastric cancer (Lee et al., 2000), metastatic head and neck squamous cell carcinomas (Di Renzo et al., 2000) and childhood hepatocellular carcinoma (Park et al., 1999). Most mutations have been found in the tyrosine kinase domain of Met and lead to autophosphorylation and constitutive activation of Met, even without HGF/SF stimulation. Moreover, mouse transgenic models using either the Met receptor (Jeffers et al., 1998; Wang et al., 2001) or the ligand or mutant of the ligand (Jakubczak et al., 1998; Takayama et al., 1997) are highly cancer prone.

We have been studying the molecular mechanisms and signaling molecules responsible for HGF/SF-Met mediated tumorigenesis and metastasis. The transcription factor Stat3 is a downstream signaling molecule activated by HGF/SF-Met signaling (Boccaccio et al., 1998) and has been reported to contribute to cell transformation induced by a diverse set of oncoproteins (Bowman et al., 2000). Stat3 is constitutively activated in many primary tumors and tumor cell lines, suggesting that signaling by this molecule is likely to be important for cell transformation. SK-LMS-1, a human cell line established from a leiomyosarcoma, expresses high levels of Met and is highly tumorigenic in athymic nu/nu mice when ectopically expressing HGF/SF (Rong et al., 1993). In this study, we show that Stat3β, a dominant negative form of Stat3 (Caldenhoven et al., 1996), has a profound inhibitory effect on HGF/SF-Met mediated growth in soft agar and tumor growth in athymic nu/nu mice, but has no obvious effect on HGF/SF-Met mediated transformed cell morphology, proliferation, invasion or branching morphogenesis.


HGF/SF induces tyrosine phosphorylation and nuclear translocation of Stat3 in SK-LMS-1 cells

The SK-LMS-1 leiomyosarcoma cell line expresses high levels of Met receptor, but only traces of HGF/SF (Rong et al., 1993; Jeffers et al., 1996) and with HGF/SF treatment, Met is rapidly tyrosine-phosphorylated. SK-LMS-1 cells stably expressing HGF/SF have been established (Jeffers et al., 1996) and in these cells, Met is constitutively phosphorylated. HGF/SF induces the tyrosine phosphorylation and nuclear translocation of Stat3 in MDCK epithelial cells (Boccaccio et al., 1998). Here, we show that HGF/SF has the same effects on Stat3 in SK-LMS-1 leiomyosarcoma cells (Figure 1). Stat3 is tyrosine phosphorylated and translocates to the nucleus in serum-starved SK-LMS-1 cells treated with HGF/SF.

Figure 1

HGF/SF induces tyrosine phosphorylation and nuclear translocation of Stat3. (a) SK-LMS-1 cells were serum starved for 16 h, followed by HGF/SF treatments (200 units/ml) for 5, 15 or 30 min. Cell extracts were prepared and subjected to Western blot analysis probed with anti-Phospho-Stat3 (Tyr705) (upper). The same blot was stripped and reprobed with anti-Stat3 antibody (lower). (b) Serum starved SK-LMS-1 cells with or without HGF treatment for 30 min were immuno-reacted with anti-Stat3 antibody, followed by stained with Texas Red-labeled secondary antibody

We tested the effects of expression of a dominant-negative form of Stat3, Stat3β, in the autocrine HGF/SF-Met SK-LMS-1 cells. Stat3β lacks of the C-terminal transactivation domain of Stat3, is a natural splice variant of Stat3, and functions as a dominant negative regulator of transcription by competing with the Stat3 DNA binding sites (Caldenhoven et al., 1996; as depicted in Figure 2a). We isolated independent clones Stat3β A, B, D and C8 stably expressing different levels of Stat3β (Figure 2B). We tested whether Stat3β overexpression affects the expression or phosphorylation of Met. We prepared cell extracts from the parental and Stat3β expressing clones and subjected them to immunoprecipitation and Western blot analysis using anti-Met antibody. As shown in Figure 2c, overexpression of Stat3β neither affected Met protein level nor its phosphorylation status in these cells.

Figure 2

Expression of dominant negative form of Stat3, Stat3β in hgf/sf-transfected SK-LMS-1 cells. (a) Schematic representation of the structures of Stat3 and Stat3β. (b) Cell extracts prepared from independent clones stably transfected with Stat3β were resolved in 10% SDS–PAGE. Western blot was probed with anti-Stat3 antibody. Arrows indicate the positions of Stat3 and Stat3β. (c) The expression and phosphorylation status of Met in SK-LMS-1 cells was not affected by overexpression of Stat3β. Equal amount of cell extracts were immunoprecipitated with anti-Met antibody. Western blot analyses were performed using anti-phosphotyrosine (upper panel) and anti-Met antibodies (lower panel)

Stat3β expression inhibits HGF/SF-Met dependent growth of SK-LMS-1 cells in soft agar

Met-expressing cells grow in soft agar in response to HGF/SF (Rahimi et al., 1998). SK-LMS-1 cells also form colonies in soft agar and display anchorage independent growth in response to HGF/SF (data not shown). Moreover, autocrine HGF/SF-Met signaling promotes anchorage independent growth of these cells in soft agar (Figure 3a). However, both clone A and clone B expressing high levels of Stat3β failed to grow in soft agar (Figure 3). A few cells from clone A and B formed large colonies (>0.1 mm in diameter) similar in size to control autocrine SK-LMS-1 cell colonies (Figure 3a), but these cells simply lost Stat3β expression (Figure 4). These data show that Stat3 activity is required for the HGF/SF-Met mediated growth in soft agar.

Figure 3

Dominant negative Stat3β inhibits HGF/SF-Met mediated anchorage independent growth of SK-LMS-1 cells in soft agar. (a) Colony growths were representatively shown after 2 weeks growing in soft agar. (b) Colony numbers were presented as colonies (>0.1 mm in diameter) per 1000 cells counted. Mean values are from three independent experiments and standard deviations are indicated by error bar

Figure 4

Losing expression of Stat3β restores transforming activity in hgf/sf-transfected SK-LMS-1 cells. (a) Three large colonies formed in soft agar plates of Stat3β overexpressing clone A were picked up and expanded in cultures (designated Stat3β-A′-1, A′-2 and A′-3). Extracts prepared from indicated cells were used for Western blot analysis detected by anti-Stat3 antibody. (b) Representative colony growths were shown after 2 weeks of growth in soft agar

Stat3β does not inhibit HGF/SF-induced in vitro invasion or branching morphogenesis of SK-LMS-1 cells

SK-LMS-1 cells branch in response to HGF/SF in three-dimensional matrigel (Jeffers et al., 1996). We tested whether Stat3β affects branching morphogenesis of SK-LMS-1 cells. As reported, in the absence of HGF/SF, control parental SK-LMS-1 cells do not branch, while autocrine HGF/SF SK-LMS-1 cells display extensive branching (Figure 5a). However, no differences were observed in branching activity between autocrine SK-LMS-1 cells with or without ectopic expression of dominant negative Stat3β (Figure 5a). The branching morphogenesis activities of autocrine HGF/SF SK-LMS-1 cells and its Stat3β derivatives are, however, HGF/SF-dependent, since HGF/SF neutralizing antibody blocks branching morphogenesis (Figure 5b).

Figure 5

Dominant negative Stat3β has no effect on branching morphogenesis of hgf/sf-transfected SK-LMS-1 cells with autocrine HGF/SF-Met signaling. (a) Indicated cell lines were subjected to branching morphogenesis assay using growth factor reduced matrigel. The photography showed the morphologies after 4 days growing in matrigel. (b) The branch forming activity of the cells was HGF/SF dependent. The branching morphogenesis of hgf/sf-transfected SK-LMS-1 cells could be inhibited by addition of anti-HGF neutralizing antibody. Photography showed the representative fields taken 3 days after treatment with the neutralizing antibody

We also tested the effect of Stat3β on the ability of the parental SK-LMS-1 cells to branch in response to HGF/SF. We generated clones of parental SK-LMS-1 cells stably expressing Stat3β (Figure 6a). The expression of Stat3β in these clones is higher than that of endogeneous Stat3 (Figure 6a) and the expression and phosphorylation status of Met receptor in response to HGF/SF shows no obvious differences between cells with or without Stat3β overexpression (Figure 6b). SK-LMS-1 cells, with or without Stat3β, display no significant differences in branching morphogenesis in response to HGF/SF (Figure 6c). We conclude that SK-LMS-1 cell branching morphogenesis does not require Stat3 activity.

Figure 6

Overexpression of Stat3β in SK-LMS-1 cells does not inhibit HGF/SF-induced branching morphogenesis. (a) Stable expression of Stat3β in SK-LMS-1 parental cells. Cell extracts were prepared from individual clones and resolved in 10% SDS–PAGE. Western blot was probed with anti-Stat3 antibody. (b) Met phosphorylation and expressions in individual clones with or without HGF treatment (200 units/ml for 10 min) were analysed by immunoprecipitation with anti-h-Met antibody, followed by Western blot using anti-Phosphotyrosine (upper) or anti-h-Met antibody (lower). (c) Branching morphogenesis assays were performed in growth factor reduced matrigel with or without addition of recombinant human HGF (200 units/ml). Representative fields from indicated cell lines were photographed 3 days after growing in the matrigel

HGF/SF-Met signaling induces invasion of parental SK-LMS-1 cells (Jeffers et al., 1996). To evaluate whether Stat3 activity is required for this phenotype, we performed in vitro invasion assays using matrigel. Differences in invasion were observed among the clones suggestive of clonal variations (Figure 7a, b). For example clone 30, expressing the highest level of Stat3β is almost as invasive as the parental SK-LMS-1 cells (Figure 7b). We conclude that Stat3 does not contribute to the in vitro invasive phenotype of SK-LMS-1 cells.

Figure 7

Stat3β has no obvious effect on the invasive phenotype of SK-LMS-1 cells. In vitro invasion assays were performed in an invasion chamber with or without HGF (200 units/ml) additions. Twenty-four hours after incubation, cells were stained and visualized under a phase contrast microscope. (a) Representative fields from indicated cell lines were photographed. (b) Invasive cells were quantified under a phase contrast microscope. Each assay was performed in triplicate

Stat3β does not affect HGF/SF induced scattering or branching morphogenesis of MDCK cells

We stably introduced Stat3β into MDCK cells (Figure 8a) and performed scattering and branching morphogenesis assays. Overexpression of Stat3β neither affected scattering activity (Figure 8b) nor branching morphogenesis in collagen I gel induced by HGF/SF (Figure 8c). These data are consistent with the SK-LMS-1 cell data and imply that Stat3 is not required for the HGF/SF-Met induced branching morphogenesis, motolity or invasion.

Figure 8

Stat3β has no effect on motility and branching morphogenesis of MDCK epithelial cells. (a) Establishing stable cell lines of MDCK with or without exogeneous expression of Stat3β. The protein level was shown by Western blot using anti-Stat3 antibody. (b) Scattering assays. MDCK cells with or without Stat3β were plated in a 96-well with or without the addition of HGF/SF. Twelve hours later, cells were stained and representative fields were photographed. (c) Branching morphogenesis assay. MDCK cells with or without Stat3β were plated in Collagen I gel with or without the addition of HGF/SF (100 units/ml). The medium was refreshed every 2 days. Representative fields were taken under a confocal microscope 6 days after plating in gel

Stat3β inhibits HGF/SF-Met signaling mediated in vivo tumorigenicity

SK-LMS-1 cells autocrine for HGF/SF are highly tumorigenic and metastatic in athymic nu/nu mice (Jeffers et al., 1996). We asked whether Stat3β affects the in vivo tumorigenicity of the stable autocrine SK-LMS-1 overexpressing Stat3β. While the autocrine cells lacking Stat3β expression grew to 2 mm in diameter in 3–4 weeks and to a diameter of 20 mm by 6 weeks, the SK-LMS-1 cells overexpressing Stat3β failed to develop tumors (Figure 9). Out of the twelve injected animals, one animal developed a 2 mm mass by week 3, however this failed to grow further and disappeared during the course of the experiment (Figure 9). A second animal developed a 2 mm tumor by week 6. The tumor continued to grow and on examination, Stat3β expression in this tumor was lost (data not shown). These results demonstrated that Stat3, as for many other oncogenes is absolutely required for the HGF/SF-Met mediated tumorigenesis.

Figure 9

Dominant negative Stat3β inhibits in vivo tumor formation of hgf/sf-transfected SK-LMS-1 cells. Stable cell lines of hgf/sf-transfected SK-LMS-1 with or without overexpression of Stat3β were subcutaneously injected onto the backs of athymic nude mice, respectively. Tumors were measured twice a week. Data shown represent the average tumor volumes from six individual mice for each group. Error bar shows the standard deviations


Aberrant HGF/SF-Met signaling has been associated with the malignant process of many solid tumors, and multiple downstream pathways induced by the HGF/SF have been implicated in proliferation, motility and invasion, and anchorage independent growth in vitro, as well as tumorigenesis and metastasis in vivo. Stat3β has no obvious effect on HGF/SF induced cell morphology changes, proliferation, invasion or branching morphogenesis, but is required for anchorage independent growth and tumorigenesis. Is Stat3β affecting pathways other than Stat3? No other target of Stat3β has been reported, but even if it does, the specific blocking of anchorage independent growth and tumorigenesis is quite remarkable and it should be possible to identify an alternative pathway(s).

Inappropriate and constitutive activation of Stat3 has been shown to be responsible for malignant tumor progression and is actuated by a diverse set of oncoproteins (Bowman et al., 2000) such as v-Src, v-Ros, v-Eyk as well as constitutively activated insulin-like growth factor I receptor tyrosine kinase (Besser et al., 1999; Bromberg et al., 1998; Turkson et al., 1998; Zong et al., 1998). Stat3 functions as a transcription factor which regulates the expressions of target genes, such as c-Myc, Bcl-xL, p21WAF1/CIP1 and cyclin D1 (Bowman et al., 2000; Bromberg and Darnell, 2000). Suppression of Stat3 function by Stat3β inhibits anchorage independent growth and tumorigenesis mediated by HGF/SF-Met signaling (Figures 3, 4 and 9). Stat3β-mediated inhibition of tumorigenesis is not due to any effect on cell proliferation because hgf/sf-transfected SK-LMS-1 cells with and without overexpression of Stat3β showed similar mitogenic activities as determined by 3H-thymidine incorporation assay (data not shown). Previously, Stat3β has been shown to suppress B16 melanoma tumor growth that leads to tumor cell apoptosis (Niu et al., 1999). We found that overexpression of Stat3β in autocrine SK-LMS-1 cells down-regulates the expression of Bcl-xL, but ectopic expression of Bcl-xL does not restore growth in soft agar (data not shown).

Stat3β has no obvious effect on in vitro invasion or branching morphogenesis of SK-LMS-1 human leiomyosarcoma cells (Figures 5 and 6) nor on MDCK cell branching morphogenesis (Figure 8). Expressing activated R-Ras in MDCK epithelial cells promotes branching tubulogenesis in collagen gel, but does not stimulate anchorage-independent growth in soft agar (Khwaja et al., 1998) and, constitutively activated Ron, a tyrosine kinase receptor structurally related to Met but with distinct activities, induces branching tubulogenesis but not growth in soft agar (Santoro et al., 1996). Both anchorage independent growth and branching morphogenesis are complex processes and therefore, it is not surprised that cells may require different combinations of activated pathways and/or quantitative differences in gene expressions to achieve such different cellular responses. However, a report that induction of epithelial branching tubulogenesis by HGF/SF depends on Stat3 pathway (Boccaccio et al., 1998) is not confirmed in our studies. Perhaps the difference is explained by the different approaches to inhibit Stat3 signaling. Moreover, we show that dominant negative Stat3β had no effect on cell motility induced by HGF/SF-Met. However, it is reported that Stat3 is essential for keratinocyte cell migration (Sano et al., 1999).

The link between HGF/SF-Met signaling and Stat3 may be direct and/or indirect. The evidence for a direct link is that Stat3 associates with the multiple docking sites of Met and thereby is activated upon HGF/SF stimulation (Boccaccio et al., 1998). However, Stat3 may also act through the c-Src kinase that can be phosphorylated and activated by HGF/SF-Met signaling (Faletto et al., 1993; Ponzetto et al., 1994; Rahimi et al., 1998). Src family kinases play very important role in Stat signaling (Reddy et al., 2000). Stat3 is constitutively activated in v-Src transformed cells, and it is conceivable that HGF/SF-Met signaling may activate Stat3 through activation of c-Src kinase. Indeed, c-Src kinase activity is required for HGF/SF-induced anchorage-independent growth of mammary carcinoma cells (Rahimi et al., 1998). Another possible way that activation can occur is through Rac GTPase. Rac is activated by and required for HGF/SF-Met induced cell spreading and dissociation in MDCK cells (Redley et al., 1995; Royal et al., 2000), although the mechanism is unclear. Importantly, dominant negative Rac1 also has been shown to inhibit the anchorage independent growth induced by the Tpr-Met in soft agar (Rodrigues et al., 1997). In Src transformation, Rac1-mediated JNK signaling is required for Stat3 serine phosphorylation and transcriptional activity induced by Src (Turkson et al., 1999). To achieve a full transcriptional activation of Stat3, not only tyrosine (705) phosphorylation and nuclear translocation must occur, but phosphorylation of serine (727) is required (Decker and Kovarik, 2000). Serine (727) phosphorylation can be stimulated by the MAPK and/or the JNK pathway (Decker and Kovarik, 2000) both of which are activated by HGF/SF-Met signaling (Rodrigues et al., 1997). In fact, activation of JNK pathway is essential for the Tpr-Met oncogene induced transformation (Rodrigues et al., 1997). Thus, multiple pathways activated by HGF/SF-Met signaling may involve cellular transformation by converging on Stat3 and leading to a fully transcriptional active form of Stat3. A defect(s) which fails to induce full activation of Stat3 may interfere with anchorage independent growth and tumorigenicity. Therefore, targeting the Stat3 pathway may provide a way to intervene in HGF/SF-Met mediated oncogenesis and tumor progression.

Materials and methods

Cell lines

The SK-LMS-1 human leiomyosarcoma cell line was obtained from American Type Culture Collection and cultured in Dullbecco's modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS). The hgf/sf-transfected SK-LMS-1 stable cell line has been described previously (Jeffers et al., 1996). MDCK epithelial cell line was cultured in DMEM supplemented with 5% FBS.

Expression plasmids, transfection and establishment ofstable cell lines

The Stat3β expression vector pSG5hStat3β has been described previously (Caldenhoven et al., 1996). Transfection was performed using FuGENE 6 transfection reagent (Roche). Briefly, cells plated in a 6-well dish were co-transfected with 1.9 μg of pSG5hStat3β and 0.1 μg of pSV2neo plasmid (Southern and Berg, 1982) carrying neomycin resistant gene or 0.1 μg of pCEP4 plasmid (Invitrogen) carrying hygromycin resistant gene. Forty-eight hours after transfection, cells were subjected to G418 (800 μg/ml; Gibco) or Hygromycin B (400 μg/ml; Gibco) selection. Two weeks later, individual clones were picked and screened for stable transfectants.

Western blot analysis

Cells growing in 6-well plates were collected in 100 μl of 1×Laemmli sample buffer (Sigma), and boiled for 10 min. Ten μl of cell extracts were resolved by 10% SDS–PAGE. The proteins were then electrotransferred onto a PVDF membrane (Invitrogen). After being blocked for 1 h with 5% of dry milk in TBS buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.1% Tween-20), blots were probed for 1 h at room temperature or overnight at 4°C by the following primary antibodies: anti-Stat3 (F-2; Santa Cruz), anti-Phospho-Stat3 (Tyr705) (New England BioLabs), anti-h-Met (C-28; Santa Cruz) or anti-Phosphotyrosine (clone 4G10; Upstate Biotechnology). Then, blots were reacted with peroxidase-conjugated antibody for 1 h followed by visualizing the proteins using ECL detection reagents (Amersham).


Cells growing in a 10 cm dish were extracted by 1 ml of lysis buffer consisting of 20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1% NP 40, 0.5% Deoxycholate, 0.1% SDS, 1 mM EDTA pH 8.0, 50 mM NaF, 1 mM Sodium Orthovanadate and supplemented with Proteinase Inhibitor Cocktail Tablets (Roche). Cell lysates were rotated at 4°C for 15 min followed by centrifuge at 15 000 g for 15 min at 4°C. After being quantified by DC protein assay (Bio-Rad), 0.5 mg of cell extracts was reacted with 10 μl of anti-h-Met (C-28) antibody overnight at 4°C. Immune complexes were collected on protein A-agarose beads (Life Technologies) and washed three times using the same buffer. Immune complexes were eluted in 2×Laemmli sample buffer (Sigma) and boiled for 5 min before being resolved by SDS–PAGE. Western blot analyses were performed as described above.


Cells were grown on a Lab-Tek Chamber Slide (Nalge Nunc International). After serum starved for 16 h, cells were treated with 200 units/ml of HGF/SF in DMEM containing 1 mg/ml of bovine serum albumin (BSA). Thirty minutes after HGF/SF treatment, cells were fixed in 3.7% formaldehyde in phosphate buffer saline (PBS). Cells were permeabilized in 0.1% Nonidet P-40 for 10 min, blocked with 5% goat serum and 1% BSA in PBS for 1 h, and incubated with anti-Stat3 (F2) antibody for 1 h. Cells were stained with Texas Red-labeled secondary antibody (Molecular Probes, Netherland), visualized and photographed under an immunofluorescence microscope.

Colony formation assay

Cells were suspended in DMEM containing 10% FBS and 0.5% SeaPlaque agarose (FMC BioProducts), and poured onto a 60 mm dish containing a semi-solid layer of 0.5% SeaPlaque agarose in DMEM supplemented with 10% FBS. Cells were fed weekly with DMEM containing 10% FBS and 0.5% SeaPlaque agarose. After 2 weeks growing in soft agar, colonies were counted and representative fields were photographed under a phase contrast microscope.

Branching morphogenesis assay

Branching morphogenesis assay for SK-LMS-1 cells was performed as described previously (Jeffers et al., 1996). For MDCK cells, type I collagen gel was used for this assay. Briefly, 10 000 cells of MDCK were suspended in 100 μl of collagen gel solution containing eight parts of Collagen I (rat tail, 3.05 mg/ml, Becton Dickinson), one part of 10×DMEM and one part of 0.5 M HEPES (pH 7.4) and plated in a 96-well dish. After gelling at 37°C for 20 min, 100 μl of DMEM-5% FBS with or without 100 units/ml of HGF/SF was added to each well. Cells were incubated at 37°C in 5% CO2 and re-fed with new medium every 2 days. Six days after incubation, cells were observed under a confocal microscope and representative fields were photographed.

In vitro invasion assay

The in vitro invasion assay was performed as previously described (Jeffers et al., 1996) using a 24-well size Invasion Chamber coated with Growth factor reduced Matrigel (Becton Dickinson).

Scattering assay

MDCK cells (5000 cells per well), were seeded onto 96-well plate and cultured for 12 h in 100 μl of DMEM and 5% FBS, with or without 10 Units/ml of HGF/SF. At the end of experiment, cells were stained with 0.2% Crystal Violet in 70% ethanol for 10 min and photographed.

In vivo tumorigenicity assay

Four-week old female athymic nude mice (Ncr nu/nu; Animal Production Area, National Cancer Institute-Frederick Cancer Research and Development Center, Frederick, MD, USA) were used for this assay. 1×105 cells were suspended in 100 μl of Hanks' Balanced Salt Solution (Life Technologies) and injected subcutaneously into the back of each mouse. Tumors were measured twice a week.


  1. Besser D, Bromberg JF, Darnell Jr JE, Hanafusa H . 1999 Mol. Cell. Biol. 19: 1401–1409

  2. Bladt F, Riethmacher D, Isenmann S, Aguzzi A, Birchmeier C . 1995 Nature 376: 768–771

  3. Boccaccio C, Ando M, Tamagnone L, Bardelli A, Michieli P, Battistini C, Comoglio PM . 1998 Nature 391: 285–288

  4. Bottaro DP, Rubin JS, Faletto DL, Chan AM, Kmiecik TE, Vande Woude GF, Aaronson SA . 1991 Science 251: 802–804

  5. Bowman T, Garcia R, Turkson J, Jove R . 2000 Oncogene 19: 2474–2488

  6. Bromberg JF, Darnell Jr JE . 2000 Oncogene 19: 2468–2473

  7. Bromberg JF, Horvath CM, Besser D, Lathem WW, Darnell Jr JE . 1998 Mol. Cell. Biol. 18: 2553–2558

  8. Bussolino F, Di RM, Ziche M, Bocchietto E, Olivero M, Naldini L, Gaudino G, Tamagnone L, Coffer A, Comoglio PM . 1992 J. Cell Biol. 119: 629–641

  9. Caldenhoven E, van Dijk TB, Solari R, Armstrong J, Raaijmakers JAM, Lammers JWJ, Koenderman L, de Groot RP . 1996 J. Biol. Chem. 271: 13221–13227

  10. Cortner J, Vande Woude GF, Rong S . 1995 The Met-HGF/SF autocrine signaling mechanism is involved in sarcomogenesis Basel/Switzerland: Birkhauser Verlag

  11. Decker T, Kovarik P . 2000 Oncogene 19: 2628–2637

  12. Di Renzo MF, Olivero M, Martone T, Maffe A, Maggiora P, Stefani AD, Valente G, Giordano S, Cortesina G, Comoglio PM . 2000 Oncogene 19: 1547–1555

  13. Faletto DL, Kaplan DR, Halverson DO, Rosen EM, Vande Woude GF . 1993 EXS 65: 107–130

  14. Fan S, Ma YX, Wang JA, Yuan RQ, Meng Q, Cao Y, Laterra JJ, Goldberg ID, Rosen EM . 2000 Oncogene 19: 2212–2223

  15. Fixman ED, Fournier TM, Kamikura DM, Naujokas MA, Park M . 1996 J. Bio. Chem. 271: 13116–13122

  16. Gherardi E, Stoker M . 1991 Cancer Cells 3: 227–232

  17. Giordano S, Di RM, Narsimhan RP, Tamagnone L, Gerbaudo EV, Chiado-Piat L, Comoglio PM . 1988 J Cell Biochem. 38: 229–236

  18. Grant DS, Kleinman HK, Goldberg ID, Bhargava MM, Nickloff BJ, Kinsella JL, Polverini P, Rosen EM . 1993 Proc. Natl. Acad. Sci. USA 90: 1937–1941

  19. Graziani A, Gramaglia D, Cantley LC, Comoglio PM . 1991 J. Biol. Chem. 266: 22087–22090

  20. Jakubczak JL, LaRochelle WJ, Merlino G . 1998 Mol. Cell. Biol. 18: 1275–1283

  21. Jeffers M, Fiscella M, Webb CP, Anver M, Koochekpour S, Vande Woude GF . 1998 Proc. Natl. Acad. Sci. USA 95: 14417–14422

  22. Jeffers M, Rong S, Vande Woude GF . 1996 Mol. Cell. Biol. 16: 1115–1125

  23. Khwaja A, Lehmann K, Marte BM, Downward J . 1998 J. Biol. Chem. 273: 18793–18801

  24. Koochekpour S, Jeffers M, Rulong S, Klineberg E, Taylor G, Hudson EA, Resau JH, Vande Woude GF . 1997 Cancer Res. 57: 5391–5398

  25. Lee JH, Han SU, Cho H, Jennings B, Gerrard B, Dean M, Schmidt L, Zbar B, Vande Woude GF . 2000 Oncogene 19: 4947–4953

  26. Maina F, Hilton MC, Ponzetto C, Davies AM, Klein R . 1997 Genes Dev. 11: 3341–3350

  27. Matsumoto K, Nakamura T . 1993 Hepatocyte Growth Factor-Scatter Factor and the Met Receptor Vol. 65: Goldberg, I.D. and Rosen, E.M. (eds) Basel, Switzerland: Birkhauser-Verlag pp. 225–250

  28. Niu G, Heller R, Catlett-Falcone R, Coppola D, Jaroszeski M, Dalton W, Jove R, Yu H . 1999 Cancer Res. 59: 5059–5063

  29. Nusrat A, Parkos CA, Bacarra AE, Godowski PJ, Delp-Archer C, Rosen EM, Madara JL . 1994 J. Clin. Invest. 93: 2056–2065

  30. Park WS, Dong SM, Kim SY, Na EY, Shin MS, Pi JH, Kim BJ, Bae JH, Hong YK, Lee KS, Lee SH, Yoo NJ, Jang JJ, Park S, Zhuang Z, Schmidt L, Zbar B, Lee JY . 1999 Cancer Res. 59: 307–310

  31. Pelicci G, Giordano S, Zhen Z, Salcini AE, Lanfrancone L, Bardelli A, Panayotou G, Waterfield MD, Ponzetto C, Pelicci PG, Comoglio PM . 1995 Oncogene 10: 1631–1638

  32. Ponzetto C, Bardelli A, Zhen Z, Maina F, dalla Zonca P, Giordano S, Graziani A, Panayotou G, Comoglio PM . 1994 Cell 77: 261–271

  33. Rahimi N, Hung W, Tremblay E, Saulnier R, Elliott B . 1998 J. Biol. Chem. 273: 33714–33721

  34. Reddy EP, Korapati A, Chaturvedi P, Rane S . 2000 Oncogene 19: 2532–2547

  35. Redley AJ, Comoglio PM, Hall A . 1995 Mol. Cell. Biol. 15: 1110–1122

  36. Rodrigues GA, Park M, Schlessinger J . 1997 EMBO J. 16: 2634–2645

  37. Rong S, Bodescot M, Blair D, Dunn J, Nakamura T, Mizuno K, Park M, Chan A, Aaronson S, Vande Woude GF . 1992 Mol. Cell. Biol. 12: 5152–5158

  38. Rong S, Jeffers M, Resau JH, Tsarfaty I, Oskarsson M, Vande Woude GF . 1993 Cancer Res. 53: 5355–5360

  39. Rong S, Segal S, Anver M, Resau JH, Vande Woude GF . 1994 Proc. Natl. Acad. Sci. USA 91: 4731–4735

  40. Royal I, Lamarche-Vane N, Lamorte L, Kaibuchi K, Park M . 2000 Mol. Cell. Biol. 11: 1709–1725

  41. Sano S, Itami S, Takeda K, Tarutani M, Yamaguchi Y, Miura H, Yoshikawa K, Akira S, Takeda J . 1999 EMBO J. 18: 4657–4668

  42. Santoro MM, Collesi C, Grasend S, Gaudino G, Comoglio PM . 1996 Mol. Cell. Biol. 16: 7072–7083

  43. Schmidt C, Bladt F, Goedecke S, Brinkmann V, Zschiesche W, Sharpe M, Gherardi E, Birchmeier C . 1995 Nature (London) 373: 699–702

  44. Schmidt L, Duh FM, Chen F, Kishida T, Glenn G, Choyke P, Scherer SW, Zhuang Z, Lubensky I, Dean M, Allilmets R, Chidambaram A, Bergerheim UR, Feltis JT, Casadevall C, Zamarron A, Bernues M, Richard S, Lips CJM, Walter MM, Tsui LC, Geil L, Orcutt ML, Stackhouse T, Lipan J, Slife L, Brauch H, Decker J, Niehans G, Hughson MD, Moch H, Storkel S, Lerman MI, Linehan WM, Zbar B . 1997 Nat. Genet. 16: 68–73

  45. Sonnenberg E, Meyer D, Weidner KM, Birchmeier C . 1993 J. Cell Biol. 123: 223–235

  46. Southern PJ, Berg P . 1982 J. Mol. Appl. Genet. 1: 327–341

  47. Stoker M, Gherardi E, Perryman M, Gray J . 1987 Nature (London) 327: 239–242

  48. Streit A, Stern CD, Thery C, Ireland GW, Aparicio S, Sharpe MJ, Gherardi E . 1995 Development 121: 813–824

  49. Takayama H, LaRochelle WJ, Sharp R, Otsuka T, Kriebel P, Anver M, Aaronson SA, Merlino G . 1997 Proc. Natl. Acad. Sci. USA 94: 701–706

  50. Turkson J, Bowman T, Garcia R, Caldenhoven E, De Groot RP, Jove R . 1998 Mol. Cell. Biol. 18: 2545–2552

  51. Turkson J, Bowman T, Adnane J, Zhang Y, Djeu JY, Sekharam M, Frank DA, Holzman LB, Wu J, Sebti S, Jove R . 1999 Mol. Cell. Biol. 19: 7519–7528

  52. Uehara Y, Minowa O, Mori C, Shiota K, Kuno J, Noda T, Kitamura N . 1995 Nature (London) 373: 702–705

  53. Vande Woude GF, Jeffers M, Cortner J, Alvord G, Tsarfaty I, Resau J . 1997 CIBA Found. Symp.: Plasminogen-Related Growth Factors Vol. 212: Bock GR and Goode JA (eds) West Sussex: John Wiley & Sons Ltd pp. 119–132

  54. Wang R, Ferrell LD, Faouzi S, Maher JJ, Bishop JM . 2001 J. Cell Biol. 153: 1023–1033

  55. Weidner KM, Di Cesare S, Sachs M, Brinkmann V, Behrens J, Birchmeier W . 1996 Nature 384: 173–176

  56. Weidner KM, Sachs M, Birchmeier W . 1993 J. Cell Biol. 121: 145–154

  57. Xiao GH, Jeffers M, Bellacosa A, Mitsuuchi Y, Vande Woude GF, Testa JR . 2001 Proc. Natl. Acad. Sci. USA 98: 247–252

  58. Zong CS, Zeng L, Jiang Y, Sadowski HB, Wang LH . 1998 J. Biol. Chem. 273: 28065–28072

Download references


We thank Brian Cao for providing anti-HGF/SF neutralizing antibody, James Resau and Eric Hudson for confocal microscope images, Dawna Dylewski, Jason Martin and Bryn Eagleson for helping on the animal study (Van Andel Research Institute). We thank the Van Andel Institute for their financial support.

Author information

Correspondence to George F Vande Woude.

Rights and permissions

Reprints and Permissions

About this article


  • HGF/SF
  • Met
  • Stat3
  • Stat3β
  • anchorage independent growth
  • tumorigenesis

Further reading