The hepatocyte growth factor/c-MET signaling axis plays an important role in tumor cell proliferation, metastasis, and tumor angiogenesis, and therefore presents as an attractive target for cancer therapy. Notably, most small-molecule c-MET inhibitors currently undergoing clinical trials are multitarget inhibitors with the unwanted inhibition of additional kinases, often accounting for undesirable toxicity. Here, we discovered SOMG-833 [3-(4-methylpiperazin-1-yl)-5-(3-nitrobenzylamino)-7-(trifluoromethyl) quinoline] as a potent and selective small-molecule c-MET inhibitor, with an average IC50 of 0.93 nM against c-MET, over 10,000-fold more potent compared with 19 tyrosine kinases, including c-MET family members and highly homologous kinases. SOMG-833 strongly suppressed c-MET–mediated signaling transduction regardless of mechanistic complexity implicated in c-MET activation, including MET gene amplification, MET gene fusion, and HGF-stimulated c-MET activation. In a panel of 24 human cancer or genetically engineered model cell lines, SOMG-833 potently inhibited c-MET–driven cell proliferation, whereas cancer cells lacking c-MET activation were markedly less sensitive (at least 15-fold) to the treatment. SOMG-833 also suppressed c-MET–mediated migration, invasion, urokinase activity, and invasive growth phenotype. In addition, inhibition of primary human umbilical vascular endothelial cell (HUVEC) proliferation and downregulation of plasma proangiogenic factor interleukin-8 secretion resulted from SOMG-833 treatment, suggesting its significant antiangiogenic properties. Together, these results led to the remarkable antitumor efficacy of SOMG-833 in vivo, as demonstrated in c-MET–dependent NIH-3T3/TPR-MET, U-87MG, and EBC-1 xenograft models. Collectively, our results suggested SOMG-833 as a promising candidate for highly selective c-MET inhibition and a powerful tool to investigate the sole role of MET kinase in cancer.
The receptor tyrosine kinase c-MET was first discovered and identified as an oncogene in a fusion form known as TPR-MET (Cooper et al., 1984). Downstream activation of mitogen-activated protein kinase and phosphoinositide-3 kinase was well established, signaling cascades responsible for the diverse process exerted by the hepatocyte growth factor (HGF)/c-MET axis, including proliferation, invasion, metastasis, and angiogenesis (Birchmeier et al., 2003; Lemmon and Schlessinger, 2010; Trusolino et al., 2010).
Several lines of evidence support the significant role of HGF/c-MET in cancer development (Gherardi et al., 2012). Aberrant c-MET activation resulting from specific genetic lesions, transcriptional upregulation, or ligand-dependent autocrine or paracrine mechanisms occurred in many types of cancers (www.vai.org/met/) (Comoglio et al., 2008). Moreover, it has been shown that the propagation of MET-dependent invasive growth signals is a general and remarkable feature of highly aggressive tumors, which spawn “pioneer” cells that move out, infiltrate adjacent tissues, and establish metastatic lesions (Comoglio and Trusolino, 2002; Trusolino and Comoglio, 2002; Boccaccio and Comoglio, 2006). This, together with the observation that c-MET is expressed by endothelial cells, and that HGF is a potent angiogenic factor, implies that inhibition of the HGF/c-MET signaling axis can potentially interfere with cancer onset and metastasis (You and McDonald, 2008). In addition, c-MET signaling is responsible for the resistance acquisition of approved therapies (Kentsis et al., 2012; Straussman et al., 2012; Wilson et al., 2012). All of these emphasize HGF/c-MET as an attractive target for cancer therapy, and several different therapeutic approaches are being clinically tested (Comoglio et al., 2008).
Notably, most c-MET inhibitors currently undergoing clinical trials are multitarget inhibitors. The unwanted inhibition of additional kinases often leads to undesirable toxicity (Broekman et al., 2011). The broad toxicity profile of multitarget kinase inhibitors also largely limited their chances in combination regimens. In contrast, highly selective c-MET inhibitors could largely avoid off-target toxicities at therapeutic doses, and favor their use in drug combinations. More importantly, in the new era of personalized medicine, where cancer care relies on validated biomarkers to identify a patient subpopulation harboring the specific molecular characteristics that is likely to benefit from a targeted therapy (Dietel and Sers, 2006; Hood and Friend, 2011; Ma, 2012), there is a significant need for targeted drugs with high specificity. As such, highly selective c-MET inhibitors represent the main direction for the development of c-MET–targeted therapy.
Here, we describe a novel, highly selective c-MET inhibitor, SOMG-833 [3-(4-methylpiperazin-1-yl)-5-(3-nitrobenzylamino)-7-(trifluoromethyl) quinolone], that showed strong potency against c-MET kinase, suppressed c-MET phosphorylation, and the downstream signaling in c-MET–overactivated cancer cell lines, as well as inhibited c-MET–dependent cellular events in tumor cells and primary endothelial cells. Moreover, SOMG-833 exhibited significant antitumor activity in several c-MET–driven xenograft models. All these findings promise SOMG-833 as a potential candidate for c-MET–driven human cancers.
Materials and Methods
SOMG-833 was synthesized at Shanghai Institute of Material Medica, Chinese Academy of Sciences (Shanghai, China) as we have reported previously (Wang et al., 2011). This compound was fully characterized and possessed a purity of 99%. The compound was prepared as a 10 mM stock solution in 100% dimethylsulfoxide and routinely stored at −80°C.
Enzyme-Linked Immunosorbent Assay Kinase Assay and ATP Competitive Assay
c-MET tyrosine kinase activity was evaluated by enzyme-linked immunosorbent assay (ELISA) as described before (He et al., 2014). Details of the procedures are described in Supplemental Materials and Methods. For ATP competitive assay, various concentrations of ATP were diluted for the kinase reaction. The results were analyzed in Lineweaver-Burk plots.
Human cancer cell lines SNU-1, SNU-5, AGS, DU-145, NCI-H661, NCI-H441, A549, H1581, Bx-PC3, HepG2, DBTRG, MCF-7, MDA-MB-231, HT-29, HCT-116, A375, and SK-MEL-28 were all purchased from American Type Culture Collection (ATCC, Manassas, VA); human cancer cell lines MKN-45 and EBC-1 were purchased from JCRB (Japanese Collection of Research Bioresources, Japan) and were routinely maintained according to ATCC’s or JCRB’s recommendations. The BaF3 cell line was purchased from DSMZ (Braunschweig, Germany). The Madin−Darby canine kidney epithelial cell line (MDCK) was kindly gifted from Prof. H. Eric Xu. SMMC-7721, BGC-803, U-87MG, and BEL-7404 cell lines were obtained from the Institute of Biochemistry and Cell Biology (Chinese Academy of Sciences); GES-1, and SPC-A4 cell lines were obtained from the Shanghai Cancer Institute, Renji Hospital and Chest Hospital, Shanghai Jiaotong University School of Medicine (Shanghai, China). Primary human umbilical vascular endothelial cells (HUVEC) were purchased from AllCells (Alameda, CA). Cells were cultured according to the suppliers’ instructions. The BaF3/TPR-MET cell line was a genetically generated BaF3 cell line that stably expressed a constitutively active oncogenic version of c-MET.
EBC-1, MKN-45, and BaF3/TPR-MET cells were cultured under regular growth conditions to the exponential growth phase and treated with SOMG-833 for 2 hours. A549, NCI-H441, MDCK, and HUVEC cells were serum starved for 24 hours and then incubated with the compound for 2 hours, and 100 ng/ml HGF (PeproTech, Rocky Hill, NJ) was added for an additional 15 minutes. Cells were then lysed in 1 × SDS sample buffer and subsequently resolved by 10% SDS-PAGE and transferred to nitrocellulose membranes. The membranes were first probed with phospho–c-MET, phospho–extracellular signal-regulated kinase (ERK), ERK, phospho-AKT, AKT (all from Cell Signaling Technology, Danvers, MA), c-MET (from Santa Cruz Biotechnology, Santa Cruz, CA), or glyceraldehyde 3-phosphate dehydrogenase (KangCheng Biotech, Shanghai, China) antibody and then with antirabbit or antimouse IgG horseradish peroxidase (Jackson ImmunoResearch Laboratories Inc., West Grove, PA). Immunoreactive proteins were detected using ECL Plus or Femto (Thermo Fisher Scientific, Waltham, MA), and images were captured with ImageQuant LAS 4000 (GE Healthcare, Chalfont St. Giles, UK).
Cell Proliferation/Survival Assays
Tumor cells were seeded in 96-well plates, 3000–8000/well, in growth media overnight and then exposed to designated concentrations of SOMG-833 for 72 hours. A sulforhodamine B (Sigma-Aldrich, St. Louis, MO) or MTT (thiazolyl blue tetrazolium bromide; Sigma-Aldrich) assay was done to determine tumor cell proliferation. HUVEC cells (passage 3) were first serum starved in complete medium (AllCells) for 24 hours and treated by SOMG-833 for 72 hours in media containing 3% bovine serum albumin and 100 ng/ml HGF. Appropriate controls were conducted [containing 100 ng/ml HGF (HGF+) or not (HGF−)]. A Cell Counting Kit-8 (Dojindo, Shanghai, China) assay was done to determine the viability of HUVEC cells. HGF-dependent proliferation inhibition % = [1 − (ODtreatment − ODHGF-/ODHGF+ − ODHGF-)] × 100%. IC50 values were calculated by concentration-response curve fitting a four-parameter method.
Cell Cycle and Apoptosis Assay
1 × 105 EBC-1 or MKN-45 cells were seeded in six-well plates (Corning, NY) overnight; the following day, cells were treated with different concentrations of SOMG-833 for 24 hours. After treatment, the cells were trypsinized, fixed in 70% ethanol, incubated in 20 ng/ml RNase and 10 ng/ml propidium iodide, and analyzed using a flow cytometer (FACS Calibur; BD, Franklin Lakes, NJ). The data were analyzed using software Modifit LT. Cell apoptosis was determined by an Annexin V-fluorescein isothiocyanate/PI Apoptosis Detection kit (Vazyme, Piscataway, NJ).
Cell Migration and Invasion
For migration assays, NCI-H441 cells suspended in serum-free Dulbecco’s modified Eagle’s medium (DMEM) at a density of 1.5 × 105 cells/ml were seeded (0.1 ml) in the plate inserts of the transwell chamber (pore size, 8 μm; Corning Life Sciences, Lowell, MA), and serum-free DMEM (0.6 ml) containing 100 ng/ml HGF (only added to the lower well) or not with designated concentrations of SOMG-833 was added. NCI-H441 cells were subsequently cultured for 24 hours. Cells that migrated to the lower wells were then fixed by 90% ethanol, stained by 0.1% crystal violet, and photographed. The crystals stained on the lower side of the well were dissolved by 100 µl of 10% acetic acid, and the absorbance of the resulting solution was measured at 600 nm using a multiwell spectrophotometer (SpetraMAX 190; Molecular Devices, Sunnyvale, CA). The group only stimulated by HGF was designed as a positive control (HGF+, 100% migrated/invasive). The relative migration/invasion was calculated as (ODtreated/ODHGF+) × 100%.
For the invasion assay, NCI-H441 cells were cultured in the top chambers containing Matrigel-coated membrane inserts (BD Biosciences, San Jose, CA). The ensuing procedure was identical to the migration assay.
Cell Scattering Assay
MDCK cells (1.5 × 103 cells per well) were plated into 96-well plates and grown overnight. Increasing concentrations of SOMG-833 and 100 ng/ml HGF were added to the appropriate wells and incubated at 37°C, 5% CO2 for 24 hours. The cells were fixed with 4% paraformaldehyde, stained by 0.5% violet purple, and photographed under a microscope.
Cell Branching Morphogenesis
Cells at a density of 20,000 cells/ml in serum-free DMEM were mixed with collagen I solution (BD Biosciences) at a proportion of 4:6 (the pH was adjusted to alkaline) and then plated at 0.1 ml/well of a 96-well culture plate and incubated for 45 minutes at 37°C, 5% CO2 to allow collagen gelling. HGF (100 ng/ml) with or without SOMG-833 at various concentrations dissolved in 100 μl of DMEM was added to each well. The medium was replaced with fresh growth medium every 2 days. Pictures were taken under a microscope after 4 days.
Urokinase Plasminogen Activator Activity Detection Assays
Urokinase plasminogen activator (uPA) activity detection was carried out according to protocols reported previously (Webb et al., 2000). Detailed procedures can be found in Supplemental Materials and Methods.
In Vivo Studies
Female nu/nu mice (4–6 weeks old) were maintained under clean room conditions and housed on particulate air–filtered ventilated racks. Animal experiments were performed according to institutional ethical guidelines of animal care.
Subcutaneous Xenograft Models in Athymic Mice.
Tumor cells at a density of 5 × 106 in 200 μl were first implanted subcutaneously into the right flank of each nude mouse and then allowed to grow to 700–800 mm3, defined as a well-developed tumor. After that, the well-developed tumors were cut into 1-mm3 fragments and transplanted subcutaneously into the right flank of nude mice using a trocar. When the tumor volume reached 100–150 mm3, the mice were randomly assigned to control and treatment groups (n = 5 per group).
Control groups were given normal saline alone, and treatment groups received SOMG-833 via intraperitoneal injection once daily. The tumor volume (TV) was calculated as follows: TV = [length (mm) × width2 (mm2)]/2. RTV (relative tumor volume) = TVDay N/TVDay 0 × 100%. Percent inhibition values were measured on the final day of the study for drug-treated compared with vehicle-treated mice and were calculated as follows: (1 – [(treated final day − treated day 0) / (control final day – control day0)]) × 100%.
Signal Transduction Studies.
At designated times (3 days) after SOMG-833 administration, mice were humanely euthanized and tumors were resected. Tumors were snap-frozen in liquid nitrogen, protein lysates were generated, and protein concentrations were determined using a bovine serum albumin assay (Thermo Fisher Scientific). The total tissue protein lysates were then conducted by Western blot for detecting the phospho–c-MET, phospho-ERK, and phospho-AKT (Cell Technology Signaling).
Cytokine Secretion Detection
The serum from EBC-1 xenograft mice was collected from vehicle and SOMG-833–treated groups on the final day (day 14) of the experiment. Cytokine secretion was detected using ELISA assays (70-E-EK1081; MultiSciences Biotech, Hangzhou, China) and a Bio-Plex Pro Human Cytokine 27-Plex Assay (M50-00031YV; Bio-Rad, Hercules, CA).
Data were presented as the mean ± S.D. (in vitro) or mean ± S.E.M. (in vivo). The two-tailed Student’s t test was performed to analyze statistical differences between groups, and P ≤ 0.05, P ≤ 0.01, P ≤ 0.001 were considered significant.
SOMG-833 Is an ATP-Competitive Inhibitor of c-MET Kinase with High Selectivity.
SOMG-833 was initially identified as a potent small-molecule inhibitor of c-MET with an IC50 of 0.93 ± 0.15 nM in a biochemical enzymatic assay (Fig. 1A) (Wang et al., 2011). We were prompted to investigate whether this potency was specifically against c-MET. A panel of 20 human kinases was profiled, including c-MET family member RON (macrophage-stimulating protein receptor) and highly homologous kinase AXL (tyrosine-protein kinase receptor UFO). In contrast to its high potency against c-MET, SOMG-833 barely inhibited other 19 tested kinases at concentrations up to 10 μM (Table 1), indicating that SOMG-833 was a selective c-MET inhibitor.
Most kinase inhibitors discovered to date are ATP competitive. To examine whether SOMG-833 functioned in this manner, we evaluated the inhibitory potency of SOMG-833 on c-MET activity using an ATP competitiveness assay. With the increasing concentration of ATP, the inhibitory activity of SOMG-833 upon c-MET kinase was decreased. In a Lineweaver-Burk plot, the different concentration curves of SOMG-833 intersected at a specific point (known as 1/Vmax) at the y-axis with a broad concentration from 0.32 to 200 nM (Fig. 1B), showing that SOMG-833 was an ATP-competitive inhibitor. Together, these data suggested that SOMG-833 is a potent and selective c-MET inhibitor that blocks c-MET kinase activity in an ATP-competitive manner.
SOMG-833 Inhibits c-MET Phosphorylation and Blocks Downstream Signals.
To confirm cellular effectiveness of SOMG-833 targeting c-MET kinase, four cell lines with different mechanisms of c-MET activation were chosen, i.e., human non–small-cell lung cancer cell line EBC-1 and gastric tumor cell line MKN-45 with MET gene amplification, a genetically engineered cell line BaF3/TPR-MET stably expressing a constitutively active oncogenic version TPR-MET, and non–small–cell lung cancer cell line A549 responsive to HGF stimulation. These cells were treated with various concentrations of SOMG-833, and c-MET signaling was examined using Western blot assays (Fig. 2A). The results showed that SOMG-833 inhibited c-MET phosphorylation in a dose-dependent manner, with a complete abolishment at 1 μM in all tested cells. Similar results were observed in EBC-1 and MKN-45 cells using immunofluorescence assay (Fig. 2B). ERK1/2 and AKT are the key downstream molecules of c-MET and play important roles in c-MET functioning. In line with suppressed c-MET phosphorylation, phospho-ERK and phospho-AKT were significantly inhibited by SOMG-833 in a dose-dependent manner in all four tested cell lines (Fig. 2A). These data showed that SOMG-833 effectively suppressed c-MET signaling in cancer cells, regardless of mechanistic complexity in c-MET activation across different cellular contexts.
SOMG Selectively Inhibits c-MET–Dependent Tumor Cell Proliferation.
Sustained c-MET signaling elevation could trigger uncontrolled cell proliferation, one of the hallmarks of cancer. We then evaluated the effect of SOMG-833 on c-MET–dependent cell proliferation. SOMG-833 significantly inhibited the proliferation of EBC-1, MKN-45, SNU-5, and BaF3/TPR-MET cell lines, whose growth was driven by activated c-MET signaling arising from MET gene amplification or TPR-MET gene fusion, with a mean IC50 value of 0.160–0.457 μM (Fig. 3A; Supplemental Table 1). The inhibitory effect of SOMG-833 on EBC-1 and MKN-45 cell proliferation was further confirmed in a colony formation assay (Fig. 3B). By expanding to a panel of cancer cell lines originating from different tissues with MET low expression or activation, SOMG-833 showed at least over 15-fold less potency (Fig. 3A). These data demonstrated that SOMG-833 specifically inhibited c-MET–dependent cancer cell growth.
SOMG-833 Inhibits c-MET–Dependent Cell Proliferation through Arresting Cells at the G1/S Phase.
c-MET inhibition is known to block cell proliferation via cell cycle arrest (Bertotti et al., 2009). To confirm whether the antiproliferative activity of SOMG-833 was associated with blockage of c-MET signaling, EBC-1 and MKN-45 cells were treated with various concentrations of SOMG-833 for 24 hours and cell-cycle distribution was analyzed. SOMG-833 induced a G1/S phase arrest in the EBC-1 cells, with 82.05% of the cell population in the G1 phase in the presence of 1 μM SOMG-833 (versus 55.78% in the control group) (Fig. 3, C and D). Similar results were recapitulated in MKN-45 cells (Fig. 3, E and F). Consistently, the cyclin-dependent kinase inhibitors p27 and p21 were significantly increased, whereas the expression of G1/S modulators CylinD1 and CyclinE1 were downregulated by SOMG-833 (Fig. 3G). Meanwhile, no obvious sub-G1 cell population was observed upon SOMG-833 treatment (Fig. 3, C and E). Treatment with SOMG-833 for up to 48 hours displayed no apparent apoptotic cells as detected by annexin V/PI dual staining (Fig. 3H), suggesting the G1/S phase arrest contributed most to the proliferation inhibition induced by SOMG-833.
SOMG-833 Potentially Inhibits c-MET–Mediated Metastasis.
HGF/c-MET axis activation promoted cell invasion and migration to allow cancer metastasis (Jeffers et al., 1996b). We then tested the effects of SOMG-833 on these processes using transwell-based migration and invasion assays. SOMG-833 inhibited HGF-induced NCI-H441 cell migration (Fig. 4, A and B). Similar results were observed in a wound-healing assay using MDCK cells (Supplemental Fig. 1). Further, SOMG-833 strongly suppressed HGF-induced NCI-H441 cell invasion (Fig. 4, C and D) under a condition where no significant viability inhibition was observed (Fig. 4I).
Cell invasion and metastasis requires degradation of surrounding ECM. HGF/scattering factor–induced uPA plays a central role in catalyzing ECM/BM (extracellular matrix/basement membrane) degradation, mainly through cleavage of plasminogen into the broader specificity protease plasmin (Jeffers et al., 1996a). The MDCK cell line was a widely used model to evaluate uPA-plasmin network expression upon HGF stimulation (Webb et al., 2000). We found that SOMG-833 inhibited the activity of plasmin cleaved by uPA upon HGF stimulation in a dose-dependent manner (Fig. 4G).
Upon HGF stimulation, c-MET induces several biologic responses that collectively give rise to a process known as invasive growth, which is pivotal in driving cancer cell invasion and metastasis (Boccaccio and Comoglio, 2006). Thus, we next examined whether SOMG-833 inhibited c-MET–associated invasive growth. In vitro, this morphogenetic program was recapitulated by stimulating cultured MDCK epithelial cells with HGF in suspension in a three-dimensional extracellular matrix (collagen) to form multicellular-branched structures, named morphogenesis (Montesano et al., 1991; Jeffers et al., 1996b). In addition, induction of epithelial cell scattering is a unique feature of HGF and is fundamental for HGF/c-MET signaling–elicited invasive growth (Birchmeier et al., 2003). We therefore chose these two representative models, cell scattering and morphogenesis, to evaluate the impact of SOMG-833 on c-MET–mediated invasive growth. MDCK cells were stimulated with 100 ng/ml HGF in the presence of various concentrations of SOMG-833. At a concentration of 3 μM, SOMG-833 showed strong inhibitory effects on cell scattering (Fig. 4E) and morphogenesis (Fig. 4F), indicating SOMG-833 inhibited HGF-induced c-MET–mediated invasive growth.
In accordance with these effects, SOMG effectively blocked phosphoinositide 3-kinase–AKT and MAPKs (mitogen-activated protein kinases) pathways (Fig. 4H), which mediated c-MET–dependent survival, invasion, and morphogenesis (Zhang and Vande Woude, 2003). Together, SOMG-833 showed its potency against c-MET–dependent migration and invasion, thus lowering the risk of tumor metastasis.
SOMG-833 Inhibits c-MET–Dependent Proliferation of HUVEC.
In addition to its crucial role in cancer cells, HGF/c-MET signaling is a potent inducer of endothelial cell growth and promoted angiogenesis (Grant et al., 1993; Abounader and Laterra, 2005; Ren et al., 2005; You and McDonald, 2008). Hence, we also assessed the antiangiogenesis potential of SOMG-833. As shown in Fig. 4J, SOMG-833 dose-dependently inhibited HGF-stimulated growth of primary HUVEC with an average IC50 of 0.1 μM. Consistently, HGF-dependent c-MET phosphorylation and its downstream signaling were potently inhibited in HUVEC cells by SOMG-833 (Fig. 4K).
SOMG-833 Shows Strong Antitumor Activity In Vivo.
To assess the in vivo antitumor efficacy of SOMG-833, three representative tumor xenograft models driven by dysregulated c-MET were chosen: a NIH-3T3/TPR-Met model, in which tumor growth was driven by constitutive active MET fusion independent of HGF stimulation; a U-87MG human glioblastoma model with HGF and c-MET comprising an autocrine loop; and an EBC-1 xenograft model specifically driven by MET amplification.
In the NIH-3T3 model, upon 14-day SOMG-833 administration, tumor growth inhibition was observed in the SOMG-833–treated group, with inhibitory rates of 79.0% (P < 0.01) and 51.0% (P < 0.05) at doses of 80 and 40 mg/kg, respectively (Fig. 5A). In the U-87MG glioblastoma model, SOMG-833 showed a similar dose-dependent inhibition of tumor growth with inhibitory rates of 73.0% (P < 0.01) and 48.0% (P < 0.05), respectively (Fig. 5B). In the EBC-1 xenograft model, SOMG-833 strongly inhibited tumor growth inhibition at doses of 40 mg/kg (56.0%, P < 0.05) and 80 mg/kg (97.9%, P < 0.001) (Fig. 5C). By examining EBC-1 tumor tissue samples collected at different time points after SOMG-833 treatment on day 3, we observed marked inhibition of intratumoral phospho–c-MET and its downstream key effectors phospho-AKT and phospho-ERK levels in the SOMG-833–treated group (Fig. 5F), suggesting the inhibition of tumor growth by the administration of SOMG-833 was associated with the blockage of c-MET signaling.
c-MET–driven cancer malignancy is mediated by its impact on both tumor cells and endothelial cells. The intratumoral mitotic index (Ki67) was assessed using immunohistochemical analysis. A significant decrease in Ki67 expression level was observed at 80 mg/kg per day of SOMG-833 in the EBC-1 xenograft models (Fig. 5D), indicating the potent inhibition of mitogenesis in vivo. Meanwhile, the contribution of antiangiogenic efficacy was assessed. SOMG-833 was evaluated for modulation of microvessel density assessed by immunostaining for CD31 (platelet endothelial cell adhesion molecule 1). At the 80 mg/kg per day dose, the SOMG-833–treated sample showed a significant reduction of CD31-positive microvessels (Fig. 5D). Moreover, we found that the levels of proangiogenic factor interleukin-8 (IL-8), which is regulated by c-MET/HGF (Yoshida et al., 1997; Li et al., 2003; Zhang et al., 2003), in the SOMG-833 (80 mg/kg)–treated group, were downregulated compared with the vehicle group using ELISA assay (Fig. 5E) and Bio-200 System (Bio-Rad Laboratories) (Supplemental Fig. 2). These results indicated that the antitumor activity of SOMG-833 is mediated by direct effects on tumor cell growth as well as antiangiogenic mechanisms. Together, SOMG-833 showed robust antitumor efficacy which was correlated with the inhibition of c-MET–mediated signaling in c-MET–dependent tumor models.
Aberrant c-MET activation has been frequently found in many human solid tumors and hematologic malignancies. Overactivation of c-MET is known to initiate tumorigenesis and promote metastasis, as well as cause therapeutic resistance (Engelman et al., 2007), underscoring the importance of developing therapeutic strategies capable of interrupting c-MET signaling (Gherardi et al., 2012). In fact, recent clinical trials of c-MET pathway-targeted agents have yielded convincing evidence for the benefit of targeting c-MET in cancer therapy, including monotherapy and combined therapy (Sequist et al., 2011; Bendell et al., 2013). Previously, we conducted a screen to discover specific c-MET inhibitors, and SOMG-833 was selected for further characterization of c-MET–targeting antitumor efficacy (Wang et al., 2011).
An important feature of SOMG-833 is its selectivity against c-MET. SOMG-833 presented an IC50 for c-MET in the nanomolar range and showed at least more than 1000-fold selectivity over a panel of 19 human kinases, including c-MET family member Ron (Wang et al., 2003). Consistently, cancer cells with low c-MET activity were markedly less sensitive (at least 15-fold) to SOMG-833 than c-MET–addicted cells. It is worth noting that most of the reported c-MET kinase inhibitors being clinically evaluated are multitarget inhibitors, often resulting in unwanted broad nonspecific toxicity (Broekman et al., 2011). Highly selective for c-MET kinase inhibition, SOMG-833 could specifically achieve the therapeutic potential of c-MET inhibition in patients harboring c-MET aberrations, and its use in biomarker-directed drug combinations in personalized medicines becomes possible. In addition, the feature of high selectivity makes SOMG-833 suitable for use as a tool inhibitor in preclinical models to dissect the role of c-MET kinase activity in cancer progression.
Activation of c-MET drives a complex morphogenetic program termed invasive growth (Boccaccio and Comoglio, 2006). Under normal conditions, invasive growth is based upon a finely tuned interplay between related phenomena, including cell proliferation, motility, ECM degradation, and survival. In transformed tissues, aberrant implementation of this interplay is responsible for cancer progression and metastasis (Comoglio and Trusolino, 2002). In our study, using a series of cell models, we were able to dissect the key biologic steps of invasive growth, including cell proliferation, scattering, migration, and invasion, and found potency of SOMG-833 against these individual steps. Moreover, SOMG-833 reversed the comprehensive three-dimensional branching morphogenesis phenotype stimulated by HGF, further confirming the strong inhibitory effect of SOMG-833 on c-MET–mediated invasive growth. All of these indicated a potential role of SOMG-833 against tumor progression and metastasis.
HGF and its receptor c-MET have been implicated in the regulation of tumor angiogenesis through multiple mechanisms (Zhang et al., 2003). In the present study, SOMG-833 showed the ability to inhibit HGF-stimulated c-MET–mediated endothelial cell survival, and to reduce microvessel density in the EBC-1 model. In addition to their reported direct role in regulating endothelial cell function, c-MET and HGF are also implicated in the regulation of secretion of angiogenic factors by epithelial and tumor cells (Zhang et al., 2003; Knowles et al., 2009; Hill et al., 2012). Zou et al. (2007) have reported that PF-2341066 (Crizotinib; Pfizer, New York, NY) could decrease IL-8 and VEGF in c-MET–dependent gastric GTL-16 and glioblastoma U-87MG models, and Torti et al. have found that reduced secretion of IL-8 may serve as an indicative biomarker responding to c-MET inhibition by pharmacological inhibitors in c-MET–dependent gastric xenografts (Torti et al., 2012). Consistently, we found SOMG-833 could downregulate serum IL-8 level in EBC-1 xenograft, indicating that antiangiogenic activity observed with SOMG-833 may be mediated by direct and indirect mechanisms. However, the serum VEGF in the vehicle group was very low in the EBC-1 xenograft (<20 pg/ml) and almost no changes were observed upon SOMG-833 treatment (data not shown), suggesting that the regulation of angiogenic factor types by the HGF/c-MET axis are not common to all tumor types.
In conclusion, we assessed the antitumor efficacy of SOMG-833, a novel selective c-MET inhibitor. SOMG-833 showed strong potency against c-MET kinase, and inhibited c-MET phosphorylation and the downstream signaling across different oncogenic forms in c-MET overactivated cancer cells. In turn, SOMG-833 inhibited c-MET–driven cellular phenotype in tumor cells and primary endothelial cells. Furthermore, SOMG-833 treatment resulted in significant antitumor activity in several c-MET–driven xenografts. In addition, intratumoral inhibition of c-MET phosphorylation, proliferation index (Ki67), and IL-8 level suggested that the concurrent antiproliferative and antiangiogenic activity of SOMG-833 accounted for its anticancer efficacy in vivo.
The authors thank Prof. H. Eric Xu (Shanghai Institute of Materia Medica) for providing the MDCK cells.
Participated in research design: Geng, Yang, Ai, Ding.
Conducted experiments: H.-t. Zhang, Wang, Chen, He, Ji.
Contributed new reagents or analytic tools: A. Zhang.
Performed data analysis: H.-t. Zhang, Wang, Chen, Ai, Geng.
Wrote or contributed to the writing of the manuscript: H.-t. Zhang, Ai, Huang, Geng.
- Received March 20, 2014.
- Accepted April 14, 2014.
H.-t.Z., L.W., and J.A. contributed equally to this work.
This work was supported by the National Program on Key Basic Research Project of China [Grant 2012CB910704]; the National Natural Science Foundation [Grants 91229205 and 81102461]; National S&T Major Projects [Grant 2012ZX09301001-007]; and China Marine Commonwealth Research Project [Grant 201005022-5].
- protein kinase B
- Dulbecco’s modified Eagle’s medium
- extracellular matrix
- enzyme-linked immunosorbent assay
- extracellular signal-regulated kinase
- hepatocyte growth factor
- human umbilical vascular endothelial cells
- Madin−Darby canine kidney epithelial cell
- thiazolyl blue tetrazolium bromide
- optical density
- propidium iodide
- (3-(4-methylpiperazin-1-yl)-5-(3-nitrobenzylamino)-7-(trifluoromethyl) quinoline)
- translocated promoter region
- tumor volume
- urokinase plasminogen activator
- vascular endothelial growth factor
- Copyright © 2014 by The American Society for Pharmacology and Experimental Therapeutics