Background
Hepatocyte growth factor (HGF) is a multifunctional cytokine that was originally described as a mesenchymal-derived factor that regulates cell growth, cell motility, morphogenesis and angiogenesis [
1‐
3] through activation of its receptor, the transmembrane tyrosine kinase encoded by the
c-Met proto-oncogene [
4]. HGF and c-Met are often co-expressed or over-expressed in a variety of human malignancies including glioblastoma and medullablastoma; and their expression level correlates with poor prognosis [
5‐
8]. The multifunctional effects of HGF:c-Met signaling in tumor cells are mediated by a network of signal transduction pathways including mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K). Paradoxically and dependent on cell context and the involvement of specific downstream effectors, both pro- and anti-apoptotic effects of HGF have been reported [
9]. It is well documented that c-Met kinase-dependent signaling is able to counteract apoptosis induced by DNA-damage through the initiation of survival signals such as the PI3K-Akt, MAPK and NFκB pathways [
10‐
13]. In addition, c-Met can bind to and sequester Fas via a kinase-independent mechanism in several types of cells, including epithelial and glioblastoma cells, and thereby prevent cell death induced by death receptor ligand [
14,
15]. On the other hand, the mechanisms by which HGF:c-Met exerts pro-apototic effects are not fully understood. It has been reported that HGF:c-Met signaling induces or sensitizes apoptotic cell death in a number of cell lines including ovarian carcinoma cell, breast carcinoma cell, mouse sarcoma cell and mouse hepatocarcinoma cell [
16‐
19]. Although the anti-apoptotic functions of the HGF:c-Met pathway appear to predominate in most biological systems, pro-apoptotic responses have been observed and could contribute to the balance between cell death and survival during the initiation and progression of certain malignancies.
Embryonic neuroectodermal malignancies such as medulloblastoma are among the most common and aggressive childhood brain tumors, and are associated with high rates of morbidity and mortality. Significant improvements in survival have been achieved by treating patients early with combinations of radiation and chemotherapy (for reviews, see [
20‐
22]). However, aggressive therapy during critical periods of CNS development results in considerable neurocognitive toxicity and durable responses in patients with recurrent medulloblastoma remain disappointing. Improving our understanding of medulloblastoma cell death and survival mechanisms and developing new strategies to overcome the inherent resistance of medulloblastoma cells to death signals could have significant impacts on survival and neurocognitive outcomes [
23,
24].
Induction of selective cancer cell death is the goal of many cancer therapies [
25]. Apoptotic cell death can be initiated by either the intrinsic mitochondrial pathway or the extrinsic death receptor pathway [
26]. Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (Apo2L/TRAIL) has the potential to initiate death receptor pathways in malignant tissues while sparing normal tissues [
27]. DR4 and DR5 are members of the TRAIL receptor superfamily sharing conserved cytoplasmic death domains essential for inducing apoptosis. Upon binding to TRAIL, DR4 and DR5 recruit the adapter molecule Fas-associated death domain (FADD), which through its death domain interacts with pro-caspases-8 and -10 to form a death-inducing signaling complex (DISC). DISC then stimulates the proteolytic activation of the caspases-8 and -10, initiating a cascade of downstream caspase activation which leads to apoptosis [
28‐
30]. Nevertheless, TRAIL does not yet offer a viable solution to brain cancer therapy since many brain tumors display complete or partial-resistance to death receptor agonists [
24,
31,
32].
Earlier work by our laboratory and others has shown that human medulloblastomas frequently express HGF, c-Met and molecular constituents of the extrinsic (death receptor) apoptosis pathway [
7,
33]. This study examines the interactions between HGF:c-Met signaling and medulloblastoma cell death/apoptosis in response to TRAIL. We found that in a representative medulloblastoma cell line, HGF promotes TRAIL-induced medulloblastoma cell death. We further examined the mechanism by which HGF exert pro-apoptotic effect. We demonstrate that the enhanced cell death response to TRAIL results from HGF-mediated DR5 over-expression and increased death-inducing signaling complex formation.
Methods
Reagents
All reagents were purchased from Sigma (St. Louis, MO) unless otherwise stated. Drugs were made in stock and diluted in cell culture medium. Recombinant human TRAIL was purchased from Millipore/Chemicon (Billerica, MA). Hepatocyte growth factor (HGF) was a gift from Genentech, Inc. (San Francisco, CA). The JNK pathway inhibitor CEP-11004 was kindly provided by Cephalon Inc. (West Chester, PA) [
34]. The c-Met kinase inhibitor PHA-665752 was kindly provided by Pfizer, Inc. (La Jolla, CA) [
35]. Caspase 3, 8, 9 inhibitors were purchased from Calbiochem (San Diego, CA).
Cell culture
Medulloblastoma cell line DAOY was originally purchased from American Type Culture Collection (Rockville, MD). DAOY cells were grown in Zinc Optimum medium (Gibco, Rockville, MD) supplemented with 10% fetal bovine serum (FBS, Cellegro, Washington, DC) and 500 μg/ml penicillin-streptomycin (Gibco). Cells were grown at 37°C in a humidified incubator with 5% CO2.
Cell viability assay
Cell viability was measured by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) assay. Cells were plated at 5,000 cells/well in 24-well tissue culture plates. After treatment, MTT was added to each well at a final concentration of 150 μg/mL, and the cells were incubated for 1–2 h at 37°C. The medium was then removed, and the cell layer was dissolved with dimethyl sulfoxide (DMSO). The formazen reaction product was quantified spectrophotometrically at 570 nm using a Spectra MAX 340pc plate reader (Molecular Devices, Sunnyvale, CA, USA). The results are expressed as the percentage of absorbance measured in control cultures after subtracting the background absorbance from all values.
Flow cytometry analysis
Apoptosis was quantified using the Annexin V-FITC apoptosis detection kit (BD Biosciences, San Diego, CA) via the manufacturer's instructions. Briefly, cells were trypsinized, pelleted by centrifugation, and resuspended in Annexin V binding buffer (150 mM NaCl, 18 mM CaCl2, 10 mM HEPES, 5 mM KCl, 1 mM MgCl2). FITC-conjugated Annexin V (1 μg/ml) and propidium iodide (PI, 50 μg/ml) were added to cells and incubated for 30 min at room temperature in the dark. To quantify cell surface DR4 and DR5, cells were harvested by trypsinization and centrifugation. The suspended cells were directly incubated with PE-conjugated anti-DR4 or anti-DR5 antibody for 30 min (1:20, EBiosciences, San Diego, CA). Non-immune mouse IgG was used as the negative control. Analyses were performed on a FACscan (Becton-Dickinson, Mountain View, CA). Data were analyzed with CellQuest software (Becton-Dickinson).
Western blot analysis
All primary antibodies used for Western blot analysis were obtained from Cell Signaling Technology (Beverly, MA) unless otherwise stated and concentrations used were according to the manufacturers' recommendations. All the secondary antibodies conjugated to horseradish peroxidase were purchased from Jackson Immunoresearch Laboratories (West Grove, PA) and were used at a 1:1000 dilution. Briefly, cells were lysed with RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% NP-40, 0.25% Na-deoxycholate) containing 1× protease and phosphatase inhibitor cocktail (Calbiochem). After sonication for 15 s, the suspensions were centrifuged at 3,000 g for 10 minutes. Protein concentrations were determined using the Coomassie Protein Assay Reagent (Pierce, Rockford, IL). Thirty micrograms of protein were subjected to 4%–20% gradient SDS-PAGE (Lonza, Pittsburgh, PA) and then transferred to nitrocellulose membranes (Amersham, Pittsburgh, PA) for 1 h. The membranes were then incubated with 5% non-fat dry milk (Carnation, Nestle Food Co., Glendale, CA) in Tris-Buffered Saline Tween-20 (TBST) for 1 h, and then incubated with primary antibody overnight at 4°C in either 5% non-fat dry milk or in 5% BSA. Membranes were subsequently rinsed in TBST, and then incubated for 1 hour at room temperature with secondary antibody conjugated with horseradish peroxidase. After incubation, membranes were rinsed, and antibody binding was detected with the enhanced chemiluminescence system (Amersham).
Immunoprecipitation
Cell pellets were treated with lysis buffer (100 mM Tris-HCL, 5 μM EDTA and 1% NP-40) containing 1× protease inhibitor Cocktail and 1× Phosphatase inhibitor cocktail. Protein concentrations were determined as above. Total protein (500 μg) were incubated with 5 μg anti-DR5 antibody (Axxora, San Diego, CA) at 4°C for 1 h. Protein A/G beads (Santa Cruz, San Diego, CA) were added to the lysate and incubated overnight at 4°C with gentle rocking. The beads were washed 5 × by repeated suspension in 500 μl lysis buffer followed by centrifugation at 10,000 g for 30 s. After the last wash, 50 μl Laemmli buffer (4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 0.004% bromphenol blue, 0.125 M Tris HCl, pH 6.8) was added to the pellet and heated at 100°C for 10 min. The samples were centrifuged again to collect supernatant and then subjected to Western blot analysis as described above.
Northern blot analysis
Northern hybridization was used to examine the changes in DR5/DR4 expression level in DAOY cells after HGF treatment. RNA was isolated using the RNeasy Mini Kit (Qiagen, Valencia, CA) following the manufacturer's directions. Oligonucleotide primers were designed as follows: for DR5 (NM_003842), a 587 bp product was amplified with primer pair 5'-caccacgaccagaaacacag/5'-gcctcctcctctgagacctt; for DR4, a 569 bp product was amplified with primer pair 5'-gctgcaaccatcaaacttca/5'-ttgtgagcattgtcctcagc. RT-PCR was conducted using DAOY RNA as a template and PCR products were subcloned in TOPO PCR cloning vectors (Invitrogen, Carlsbad, CA) and sequenced prior to use. Northern blot analysis was performed with
32P-labeled cDNA probes as previously reported [
36]. Radioactivity was quantified by densitometry and by phosphorimaging using the Bio-Imaging analyser BAS 2500 (Fuji Medical Systems; Stamford, CT). All blots were stripped in 1 × saline sodium citrate buffer (SSC)/0.1% sodium dodecyl sulfate (SDS) maintained at a temperature of 85°C for 10 min and then rehybridized with cDNA specific for 28S rRNA. Results are expressed relative to 28S rRNA.
Quantitative reverse transcription-PCR (RT-PCR)
Total RNA was extracted from snap-frozen tumor tissue using Trizol reagent (Invitrogen) via manufacturer's instructions. RNA was then treated with DNase and further purified using the RNeasy protocol (Qiagen). RT-PCR for tumor necrosis factor receptor superfamily, member 10b (DR5) was done as previously described [
37], with all reactions normalized to actin (Applied Biosystems, Foster City, CA). Commercially available Assay on Demand TaqMan primers and probes were used to measure mRNA for DR5 (catalog number 4331182). Serial dilution of total RNA from DAOY was used to generate standard curves. The expression of c-Met and DR5 in all clinical samples was calculated in relation to these. Each quantitative RT-PCR reaction was done in triplicate and error bars represent SE.
Statistical analysis
Statistical analyses consist of one-way ANOVA followed by the Tukey multiple comparison tests using Graphpad Prism (GraphPad, San Diego, CA). Log Pearson analysis (Graphpad Prism) was used to determine the correlation between c-Met and DR5 expression in human embryonal CNS tumors. All experiments reported represent at least three independent replications. Data are represented as mean value ± standard error of mean (SE).
Discussion
We show that HGF promotes TRAIL-induced cell death in the human medulloblastoma DAOY cell line. We implicate a mechanism by which activating the c-Met tyrosine kinase enhances DR5 death receptor expression, resulting in enhanced DISC formation and downstream effector caspase-dependent apoptosis in response to TRAIL. This is, to the best of our knowledge, the first study demonstrating up-regulation of DR5 by the HGF:c-Met signaling pathway. We also identify a relationship between c-Met and DR5 expression level in clinical brain tumor specimens, suggesting that our in vitro findings may be translatable to a subset of human embryonal CNS tumors.
HGF is well known for its ability to protect tumor cell death triggered by various DNA-damaging stimuli including gamma-radiation and chemotherapeutic agents [
10‐
13]. We reported previously that HGF protects DAOY medulloblastoma cells against chemotherapy-induced apoptosis [
7]. MacDonald and colleagues have compared the gene expression patterns of medulloblastoma cell lines to medulloblastoma clinical specimens and concluded that DAOY cells most closely mimic metastatic medulloblastoma, supporting the potential relevance of our in vitro results [
40]. The potential that our apoptosis-promoting results might be more broadly applicable to human medulloblastoma is further supported by the correlation between c-Met and DR5 expression level in human embryonal CNS tumors. The cell specific molecular determinants of pro- and anti-apoptosis responses to HGF:c-Met pathway activation are of interest and will require further investigation.
The second messenger pathways downstream of c-Met activation are multifunctional and can include cytoprotection, proliferation or death promotion [
11,
14,
36,
38]. The PI3K-AKT pathway, which is potently activated by HGF, mediates cell protection in a large number of in vitro models. In DAOY cells, the overall effect of c-Met activation was to enhance TRAIL-induced death. We used the PI3K specific inhibitors LY294002 and wortmannin to specifically examine the role of PI3K-AKT pathway activity in the sensitivity of HGF-treated DAOY cells to TRAIL. While not statistically significant, we found that both inhibitors increased TRAIL-induced cell death in HGF-treated cells. Thus, it is possible that PI3K-AKT pathway activation by HGF:c-Met signaling does have cytoprotective actions against in TRAIL-induced DAOY cell apoptosis even though the predominant effect of c-Met activation is death promoting under our experimental conditions. Things become more complicated when the cells were pre-treated with the MAPK specific inhibitor PD98059. Similar as PI3K-AKT, MAPK play an important role in cytoprotection (6, 7). Our Northern blot analysis data showed that MAPK may also mediate the DR5 up-regulation induced by HGF. The two aspects of MAPK may counteract and this may explain that the net effect of PD98059 is neutral. The JNK (c-jun NH2-terminal kinase), a well known cell death promoting signaling pathway, was found to be activated by HGF and to mediate HGF-induced cell death in other tumor models [
38]. We found that in DAOY cells, JNK was transiently activated by HGF, peaking within 1 hour, and then quickly returned to base line. It has been postulated that sustained JNK activation is required to activate death-promoting downstream targets such as Bim and Bak [
41]. The fact that JNK was only transiently activated by HGF is consistent with at most a minor role of JNK in sensitizing DAOY cells to TRAIL induced apoptosis. The fact that the JNK pathway specific inhibitor CEP-11004 failed to reverse the death promoting effect of HGF further supports the notion that sustained JNK activation is required to promote apoptosis.
C-Met activation by HGF can inhibit Fas-induced hepatocyte apoptosis by preventing the degradation of FLIP
L
in response to activated Fas [
15]. In contrast to this, we found that c-Met activation does not modulate FLIP
L
levels in DAOY cells treated with TRAIL. Whether or not HGF affects FLIP function via other mechanisms will require further investigation. Complex formation between c-Met and the death receptor Fas has been found in other cell lines [
14,
19]. In these reports, the interaction between c-Met and Fas was found to inhibit DISC formation and thereby protected cells from apoptosis. Based on these reports, it is possible that HGF might dissociate c-Met from TRAIL receptors and thereby promote TRAIL-induced cell death [
19]. However, co-immunoprecipitation experiments did not detect complexes with c-Met and DR5, the predominant TRAIL receptor expressed by DAOY cells. Instead, HGF was found to upregulate DR5 expression and increase the formation of death-inducing signaling complexes containing DR5 and caspase-8. Overall, our findings suggest that DR5 enhancement is the predominant mechanism by which HGF potentiates TRAIL-induced cell death in DAOY cells.
Conclusion
In conclusion, we have demonstrated for the first time that c-Met kinase activity can up-regulate death receptor expression and thereby sensitize tumor cells to the death receptor ligand TRAIL. Our analysis of clinical specimens suggests a correlation between c-Met and DR5 expression level in a subset of human embryonal CNS malignancies. These findings extend the notion that the c-Met receptor tyrosine kinase can have dual anti-apoptotic and pro-apoptotic actions in a common pediatric brain malignancy. Our findings suggest that death receptor activation may provide a route to overcome chemo/radio-resistant cells in subsets of embryonal CNS malignancies that express an activated c-Met pathway.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
YL conducted most of the experiments in this study. XF contributed Fig.
6. CRG contributed part of Fig.
1. JL was responsible for the overall design of the study, and edited the manuscript. SX initiated the study and drafted the manuscript. All authors read and approved the manuscript.