Impact of p53 Status on Radiosensitization of Tumor Cells by MET Inhibition–Associated Checkpoint Abrogation Free

The receptor tyrosine kinase (RTK) MET for HGF is involved in various essential biologic functions, mainly in embryogenesis, tissue repair, and regeneration (1, 2). Altered MET expression has been largely implicated in cancer initiation, development, and poor clinical prognosis (3). Deregulation of the MET/HGF system, which arises primarily due to the MET gene amplification and MET protein overexpression, results in ligand-independent persistent activation of the receptor and its downstream signaling pathways that include the MAPK, the PI3K–AKT, and the STAT signaling cascades (4). Consequently, the aberrant receptor function leads to uncontrolled cell proliferation, local invasion, tumor-related angiogenesis, and metastasis (5, 6). Blocking of the MET kinase activity by selective anti-MET monoclonal antibodies or small-molecule inhibitors has demonstrated to potentiate antitumor effects on many types of human cancer models (5). In this respect, MET has emerged as a molecular target for cancer treatment and several drugs directed against the MET/HGF system are currently under clinical evaluation (7).

Emerging data from recent studies suggest a potential crosstalk between the MET signaling and the cellular response to DNA damage that is associated with tumor cell resistance to DNA-damaging agents such as ionizing radiation (IR) or radiomimetic drugs (8), indicating that the MET/HGF system potentially protects tumors from these anticancer treatments. For instance, cancer cells display a MET/HGF–mediated resistance to apoptosis following treatment with various DNA-damaging agents including IR (9, 10). As reported by Fan and colleagues in 2007, one of the resistance mechanisms that protect the cell from the infliction of DNA damage might be regulated by the RAS protein–induced NF-κB activation through HGF, thus stimulating antiapoptotic signaling and cell survival (11). Furthermore, the same group has also shown that tumor cells markedly enhance the repair of the DNA strand breaks following IR and the topoisomerase IIα inhibitor adriamycin treatment in MET/HGF–dependent fashion (12). On the basis of these reports, targeting MET/HGF in combination with chemotherapeutic agents has evolved to a promising strategy to overcome tumor chemo- and radioresistance (13, 14). Our recent findings demonstrated that inhibiting the MET receptor using the tyrosine kinase small-molecule inhibitor PHA665752 leads to the disruption of the ATR/CHK1/CDC25 checkpoint pathway and impairs the DNA repair via homologous recombination (HR) in MET-addicted cancer cells upon DNA damage (15, 16), supporting the principle of tumor cell sensitization to DNA damage via MET inhibition.

Activation of checkpoint pathways is an early cellular response to DNA damage where CHK1 and CHK2 checkpoint kinases that are phosphorylated by the master DNA damage response serine/threonine kinases ATM and ATR, play central roles (17, 18). Following DNA damage, CHK1 facilitates the intra-S and G₂–M cell-cycle arrests, while CHK2 transduces the G₁–S checkpoint signaling to the key regulatory component p53 to prevent further cell-cycle progression (19, 20). As human malignant tumors frequently harbor inactive p53 due to p53 gene mutations as well as other alterations in the p53 pathways, cells lack the G₁–S cell-cycle checkpoint and thus strongly depend on the remaining G₂–M as well as intra-S checkpoint pathways in response to DNA damage (21, 22). On the basis of the phenomenon of synthetic lethality, targeting CHK1 therefore represents a promising concept to sensitize tumors to cytotoxic agents in case of compromised p53 activity (17). Several potent CHK1 inhibitors such as AZD7762, CHIR-124, or SAR-020106 have been reported to enhance the cytotoxicity of DNA-damaging agents in p53-defective tumor cells by the abrogation of the cell-cycle arrest and premature entry into mitosis (23–28).

In the current work, we investigated the effect of MET inhibition-associated sensitization to DNA damage in p53-deficient tumor cells. We have previously reported that MET inhibition interferes with CHK1-dependent cell-cycle checkpoint and thus leads to (post-) mitotic cell death upon DNA damage in wild-type p53 cells. We hypothesized that DNA damage upon MET inhibition in p53-deficient cells leads to accumulation of damaged cells in mitotic phases and consequently results in higher cytotoxicity, based on the synthetic modality approach (15). To test this hypothesis, we made use of MET-overexpressing human cancer cell lines (bearing wild-type p53) that were infected with a shRNA lentiviral vector against p53 to generate p53-proficient and p53-depleted isogenic cell lines. Following combined treatment of a novel selective MET small-molecule inhibitor EMD-1214063 and IR, various biologic endpoints, such as viability, cytotoxicity, apoptosis, and cell-cycle distribution were evaluated in p53 knockdown cells along with their corresponding parental variants. Overall, the data suggest that p53-deficient cells are more susceptible to EMD-1214063 and IR combined treatment than the wild-type p53 line.

Human gastric carcinoma cells GTL-16, KATO II, and human non–small cell lung carcinoma (NSCLC) line H1993 were obtained from Dr. Paolo Comoglio (University of Torino, Torino, Italy), Dr. Morag Park (McGill University, Montreal, Quebec, Canada), and from Dr. Sunny Zachariah (University of Texas Southwestern Medical Center, Dallas, Texas), respectively. Human lung squamous cell carcinoma line EBC-1 and human gastric carcinoma cells MKN-45 and SNU-5 were provided by Dr. Silvia Giordano (University of Torino, Torino, Italy).

Stable transduction with lentiviral vectors for scrambled control and p53 short hairpin RNA (shRNA) were described previously (29).

EMD-1214063 [3-(1-(3-(5-(1-Methylpiperidin-4-ylmethoxy)-pyrimidin-2-yl)-benzyl)-1,6-dihydro-6-oxo-pyridazin-3-yl)-benzonitrile); Merck] and CHIR-124 [(S)-3-(1H-benzo[d]imidazol-2-yl)-6-chloro-4-(quinuclidin-3-ylamino)quinolin-2(1H)-one; Selleck Chem] were dissolved in DMSO and used at indicated concentrations.

Phospho-MET (Tyr1234/35; #3126), MET (#4560), phospho-CHK1 (Ser345; #2348), CHK1 (#2360), phospho-histone H3 (Ser10) AlexaFluor488 conjugate (#3465), phospho-histone H2AX (Ser139) AlexaFluor647 conjugate (#9720), phospho-p53 (Ser15; #9284), and p53 (#9282) antibodies were purchased from Cell Signaling Technology. Phospho-histone H2AX (#05-636), p21 (#05-655) and β-actin (#MAB1501) antibodies were obtained from Millipore Corporation.

DNA-binding activity of p53 was measured by Human Total p53 DuoSet IC sandwich ELISA (R&D Systems) according to manufacturer's instruction.

The cells were transiently transfected with p21 reporter construct (WWP-Luc) using TransIT-2020 transfection reagent (Mirus Bio) and irradiated after 48 hours. The luciferase activity was measured at indicated post-IR time by ONE-Glo Luciferase Assay System (Promega). The Renilla luciferase control plasmid (pRL-TK) was cotransfected and detected using Renilla-Glo (Promega) for the normalization.

Cell death and viability were assessed by the Live/Dead Assay Kit (Molecular Probes). Briefly, cells were stained with green-fluorescent calcein-AM (viable cells) and red-fluorescent ethidium homodimer-1 (dead cells) to distinguish the living cells from the dead cells. Images were captured and analyzed under the fluorescence microscope (Leica). Cellular viability and proliferation were determined by the tetrazolium-based metabolic assay [Cell Proliferation Kit II (XTT), Roche]. The absorbance was measured at 490 nm on a microplate spectrophotometer. The cytotoxicity was quantified by the CellTox Green (Promega) according to manufacturer's instruction.

Cells were grown to 60% to 70% confluency in six well plates and subsequently treated by EMD-1214063 (50 nmol/L) or vehicle (DMSO). After 16 hours of treatment, cells were irradiated (4 Gy) and harvested 2 hours after irradiation. Cells were then reseeded into 60 mm dishes (600 cells per dish) and cultured in drug-free medium for up to 13 days. Colonies were fixed and stained in 0.5% crystal violet dissolved in methanol and acetic acid. Colonies were scored using the “Analyze particles” plugin (ImageJ software, imagej.nih.gov/ij/).

Apoptosis level was measured based on the caspase-3 activity by utilizing the Ac-DEVD-AMC–specific caspase-3 substrate (Calbiochem). Upon addition of the caspase-3 substrate to the cell lysates, the fluorescence was detected at 460 nm on a microplate reader (Tecan). The results were normalized to the total protein concentration determined by Bradford Protein Assay.

Protein concentration of the whole-cell extracts was determined as described previously (30). Fifty micrograms of lysates were resolved by SDS-PAGE, transferred to PVDF membrane (Life Technologies), and probed with primary antibody. Proteins were detected by enhanced chemiluminescence reagent (Amersham).

Total RNA was extracted from cultured cells using TRIzol reagent following the manufacturer's instructions (Roche). Reverse transcription of mRNA was performed using the Omniscript RT Kit (Qiagen). Quantitative PCR of CHK1 was done in a 7900HT Fast Real-time PCR System (Applied Biosystems) using a TaqMan assay (Applied Biosystems). The mean C_t was determined from triplicate experiments. mRNA levels were normalized to the level obtained for HMBS. Changes in expression were calculated using the ΔΔC_t method.

The cell-cycle distribution and the population of damaged and mitotic cells were determined by flow cytometry. Briefly, cells were fixed, stained with p-histone H3 (AlexaFluor488 conjugate), γH2AX (AlexaFluor647 conjugate), and propidium iodide (PI), and acquired on a flow cytometer (LSR II, BD). Cell cycle was evaluated by Dean/Jett/Fox algorithm using FlowJo Analysis Software.

Imaging flow cytometry was used to analyze the genotoxicity in mitotic population by identifying micronuclei-containing cells. Briefly, fixed cells were labeled with phospho-histone H3 (AlexaFluor488 conjugate), phospho-histone H2AX (AlexaFluor647 conjugate), and DNA (DAPI). The samples were acquired on the ImageStreamX Mark II (Amnis, Millipore) and analyzed using IDEAS analysis software (Amnis, Millipore).

The cBioPortal for Cancer Genomics (http://www.cbioportal.org/public-portal/cross_cancer.do; refs. 31, 32) was used to obtain information on co-occurrence of TP53 mutations and diverse MET alterations previously shown to result MET aberrant signaling (gene amplification, mRNA overexpression, and phosphorylation status) in different types of cancer.

Data were evaluated using two-tailored unpaired Student t test, P values of <0.05 were considered as significant (*, <0.05; **, <0.005; ***, <0.0005). Multiplicative systematic errors were eliminated for values obtained in cytotoxicity and caspase-3 assays as previously described (33).

Previous work from our laboratory demonstrated the inhibition of the ATR-CHK1 signaling pathway in human gastric adenocarcinoma MET-overexpressing cell line GTL-16 as well as in NIH-3T3 mouse fibroblasts stably expressing the activating MET-mutated variant M1268T following treatment with anti-MET small-molecule PHA665752 (15). Here, an analogous decrease in the basal phosphorylation level of CHK1 on Ser345 (p-CHK1) was observed using a different selective MET tyrosine kinase inhibitor, EMD-1214063, in two MET-overexpressing cell lines, namely GTL-16 and the human NSCLC line EBC-1 (Fig. 1A). Moreover, the increase in CHK1 phosphorylation upon IR was also markedly impaired when this MET inhibitor was added prior to DNA damage. This effect of EMD-1214063 was also present in irradiated human gastric cancer cell lines KATO II and SNU-5, both of which overexpress the MET receptor and are classified as MET-dependent (34). Interestingly, prolonged MET inhibition by EMD-1214063 was accompanied also with a decrease in total amount of the CHK1 protein but not a decrease in its mRNA, indicating a potential MET signaling–related CHK1 regulation on a translational level (Fig. 1B). Because of the cross-talk and redundancy between CHK1 and CHK2 that have been reported in various cellular systems, we also investigated CHK2 activation status in GTL-16 and EBC-1 cells upon MET inhibition. However, unlike CHK1, the cellular levels and phosphorylation of CHK2 do not seem to be affected by EMD-1214063 (Supplementary Fig. S1).

As CHK1 plays a critical role in executing intra-S and G₂–M cell-cycle arrests upon DNA damage and MET inhibition compromises this checkpoint, we hypothesized whether cells deprived of the G₁ cell-cycle checkpoint (i.e., cancer cells lacking functional p53) would exhibit higher levels of DNA damage upon IR in combination with MET inhibition as compared with cells harboring intact p53 (20). To answer this question, various MET-dependent cell lines were treated with EMD-1214063. In these cells, MET phosphorylation at Tyr1234/1235 (Fig. 1A and C) as well as cell viability and proliferation remarkably decreased regardless of their p53 status (Supplementary Fig. S2A). Initially, we have shown that the increase of the DNA double-strand break marker Ser139 phosphorylated histone H2AX (γH2AX) upon irradiation, MET inhibition, and the combined treatment was also present across all these cell lines in a similar fashion as reported previously (Fig. 1A; ref. 15). This outcome indicates a slight tendency towards higher sensitivity to MET inhibition from higher γH2AX levels in p53-mutated (EBC-1 and SNU-5) in comparison with p53-proficient (GTL-16 and KATO II) cell lines (Supplementary Fig. S2B). To provide a solid link and to thoroughly investigate possible relation between the γH2AX levels and cellular p53 status, four different cell lines with deregulated MET activity, GTL-16, MKN-45, H1993, and EBC-1, were infected with shRNA against p53 through lentiviral delivery to generate isogenic pairs of scramble control (shC) and p53 knockdown (shp53) lines. GTL-16 and MKN-45 shC cells harboring wild-type p53 (35, 36) showed an upregulation in Ser15 phosphorylation of p53 (p-p53) as well as in total p53 levels following irradiation, while their shp53 counterparts displayed significantly lower levels of p-p53, total p53, and p21 in all the conditions (Fig. 1C). In addition, the reduction of p21 transactivation upon IR in GTL-16 shp53 compared with the parental line shown by the reporter gene assay further indicates the efficiency of p53 knockdown (Supplementary Fig. S2C). Moreover, an overall higher expression of γH2AX was detected in GTL-16 and MKN-45 shp53 cells as compared with their corresponding control lines (Fig. 1C). Importantly, this effect was not found in H1993 and EBC-1 isogenic pairs whose parental lines harbor mutated p53 (Fig. 1C; refs. 37, 38).

We further evaluated the impact of MET inhibition and irradiation on cell-cycle distribution in GTL-16 shC and shp53 cells after 72 hours. While minor differences in the counts of S-phase cells were found both between the various conditions and the two cell lines, an overall decrease in G₁ population was detected in p53 knockdown as compared with control cells (Fig. 2A). This effect was particularly obvious when cells were exposed to irradiation alone or in combination with EMD-1214063. On the other hand, p53 knockdown cells displayed an elevated G₂–M-population as compared with control cells. Similar results were obtained using the selective CHK1 small-molecule inhibitor CHIR-124 as a single treatment and in combination with IR in the GTL-16 isogenic pair (Supplementary Fig. S3A). This indicates that the cells accumulate either in G₂- or in M-phase upon concomitant abrogation of both checkpoint regulators CHK1 and p53. It is important to note that CHIR-124 was used as an example of a direct CHK1 inhibition for comparison with the effects of the MET inhibitor EMD-1214063 that presumably also inhibits CHK1 signaling, albeit indirectly via inhibition of MET and downstream signaling.

The levels of histone H3 phosphorylation (p-HH3), a well-established marker for mitosis, were determined to quantify the cells in mitotic phase of the cell cycle. We evaluated the time-dependent counts of p-HH3- as well as γH2AX-positive cells following dual treatment with EMD-1214063 and irradiation for up to 72 hours. While the number of p-HH3–positive cells decreased after 2 hours, γH2AX-positive cells were highly abundant immediately after IR for a short period of time (Fig. 2B and C). Both DNA damage and mitotic marker–positive populations gradually increased from 24 hours on, in which shp53 line revealed consistently higher percentages of mitotic and γH2AX-positive cells as compared with the shC cells. Importantly, the observed increase in number of cells bearing DNA damage seems to be specifically related to the combined treatment as no such effect was present after 72 hours in cells treated by MET inhibition or IR alone (Fig. 2C).

To test whether cells with depleted p53 are more susceptible to cell death following EMD-1214063 and IR as compared with their p53-proficient counterparts, we aimed at simultaneous detection of viable and dead cell populations using Live/Dead Assay (Molecular Probes). Living cells, which were determined by the intracellular esterase activity, were monitored using fluorescence microscope together with dead cells that were stained with EthD-1. Our data show decreased number of viable and higher counts of EthD-1–positive GTL-16 shp53 cells treated with the MET inhibitor EMD-1214063 as compared with their corresponding p53-proficient control cell line after 72 hours (Fig. 3A). These effects are further enhanced when EMD-1214063 treatment is combined with irradiation. In addition, the cytotoxicity mediated by these treatments was verified using CellTox Green Assay (Promega) that utilizes asymmetric cyanine-based dye that fluoresces upon binding to the dead cell DNA. Similar to the aforementioned results, the shp53 GTL-16 cells exhibited significantly higher levels of dead cells throughout the conditions, especially in the combined EMD-1214063/IR treatment (Fig. 3B). These observations have been further confirmed by significantly decreased clonogenic cell survival of shp53 GTL-16 as compared with p53-proficient cells when combining IR and MET inhibition (Fig. 3C). Similarly to the effects of EMD-1214063, the combination of CHIR-124 and IR synergistically increased the number of dead cells in shp53 GTL-16 cells, revealing more than 2-fold higher toxicity in p53 knockdown than in the control cells (Supplementary Fig. S3B).

We further raised the question whether the increase in cell death observed in shp53 cells as compared with shC cells upon the EMD-1214063/IR treatment is nevertheless of apoptotic nature. We therefore assessed the apoptotic response as one of the major cell death mechanisms by measuring the enzymatic activity of caspase-3. While irradiation slightly induced apoptosis, EMD-1214063 triggered more than 20-fold higher caspase-3 activity and this in both GTL-16 shC and shp53 cells (Fig. 3D). Furthermore, combined EMD-1214063/IR treatment synergistically increased the apoptotic event, in which the p53 knockdown cells showed similar levels of caspase-3 activity as the p53-proficient line. These observations indicate that although MET inhibition does induce apoptosis, this mode of cell death may not provide necessarily the rationale for increased cytotoxicity in cells with compromised p53 as compared with their isogenic cell line.

To shed more light on cell death mechanisms operative in p53 knockdown cells following combined MET inhibitor and IR treatment, morphologic and phenotypic analyses were performed in GTL-16 control and p53 knockdown lines upon the dual treatment. The γH2AX expression and DNA content were then evaluated in p-HH3–positive population that represents cells in mitotic phase. While proper cell divisions were observed both in untreated shC and shp53 cells, EMD-1214063/IR-treated population revealed aberrant mitosis accompanied by increased γH2AX expression, which was detected predominantly in the p53 knockdown cells (Fig. 4A). Moreover, the DNA staining implicates an indication of micronuclei formation in these cells suggesting the genotoxicity of the treatment. The quantification of polyploid (>4n) cells as well as the sub-G₁ fraction, which represents DNA fragmentation, substantiates enhanced susceptibility of shp53 cells to irradiation in combination with MET inhibition, suggesting the acquisition of mitotic aberrations as a potential cause of cell death following mitosis (Fig. 4B). In support of these findings, a decrease in cytotoxicity upon the combined EMD-1214063/IR treatment can be observed in shp53 cells once their mitotic progression is pharmacologically blocked using either nocodazole or the Polo-like kinase 1 inhibitor BI 2536 (Supplementary Fig. S4).

To assess whether our findings are congruent with the molecular alterations found in human tumors, we analyzed the data sets from The Cancer Genome Atlas available at the cBioPortal for Cancer Genomics (http://www.cbioportal.org/public-portal/cross_cancer.do; refs. 31, 32) to match the genetic status of TP53 with copy number alteration, mRNA expression, and reverse phase protein arrays (RPPA) data for MET in 5 different tumor types (Table 1). In this selected group of tumors, there was a statistically significant tendency towards co-occurrence of alterations in TP53 and MET [as represented by odds ratio (OR) values significantly superior to 0]. MET gene amplification was generally uncommon (0%–4.65%), but tumors with MET amplification harbored TP53 mutations in most lung and gastric adenocarcinomas (Table 1). mRNA overexpression was variable across tumoral entities (2.06%–16.09%), and co-occurrence of TP53 mutations in such tumors was equally variable but tended to be higher than in the case of gene amplification. Finally, RPPA upregulation was relatively rare, and the levels of TP53 mutations co-occurrence ranged between 6.67% and 28.57% (Table 1).

Our previous work demonstrated a MET targeting–associated DNA damage response in MET-addicted cell lines by assessing increased γH2AX levels following exposure to the anti-MET–selective drug PHA665752 (15). Here, we report higher endogenous expression of γH2AX in p53 knockdown cells as well as in all tested p53-mutated lines in comparison with the wild-type p53 cells and elevated γH2AX in compromised p53 cells upon treatment with EMD-1214063, indicating genetic instability that potentially derives from the absence of the p53 function.

The phenotypic characteristics of elevated γH2AX background expression in p53 mutants as compared with the wild-type p53 cancer cells from a selected panel of NCI-60 tumor lines was previously reported (39). Although the retained γH2AX foci have been correlated with ultimate cell death that may arise from repair deficiency, replication stress, and telomere dysfunction, these expression levels may not be sufficient to predict the sensitivity towards radiation or chemotherapy (40).

Several studies have suggested the direct link between MET-driven tumors and their p53 status. For instance, a potential crosstalk between p53 and oncogenic MET activity was described by Furlan and colleagues, suggesting c-Abl as a signaling node that interconnects the MET and the p53 signaling pathway, in which wild-type p53 may contribute to tumorigenesis in interplay with oncogenic RTK system (36). However, most studies demonstrate that wild-type p53 exerts tumor-suppressing functions whereas compromised p53 is oncogenic. It has been proposed that esophageal epithelial cells harboring p53 mutation (R175H) gain oncogenic function and contribute to increased cell invasion via elevated MET activation (41). The high activity of the MET receptor and its association with enhanced motility, invasion, and scattering was correlated with the p53-dependent regulation of MET expression (42). Loss of wild-type p53 function due to mutations was shown to result in MET overexpression, increased MET signaling, receptor recycling, and consequently in progression to metastatic phenotypes. Moreover, the p53-mediated MET expression is facilitated both in miR-34a-dependent and -independent manner via p53 (43–45). Taken together, the aggressive tumorigenic effect that associates with mutated p53, which has been also correlated with potential resistance to chemotherapeutic drugs, should be approached to improve cancer therapy outcome (46). As suggested by the current study, in the case of at least some MET-driven tumors, targeting of MET especially in p53-mutated background, may provide therapeutic benefit when a MET inhibitor is utilized concurrently among others, as a potential checkpoint abrogator.

As one of the essential steps in cellular DNA damage response, the activation of CHK1 prevents entry into mitosis in the presence of DNA lesions either at intra-S and/or at G₂–M transition. Thus, the concept of targeting CHK1 in tumors harboring deficient p53 for selective sensitization to radiation or to chemotherapeutic agents has been validated over the last decade using numerous CHK1 inhibitors (23–28, 47). Inhibition of CHK1 in various cancer types harboring p53 mutation including breast cancer, multiple myeloma, colon carcinoma, pancreatic cancer, and brain glioblastoma resulted in cell death mainly via mitotic catastrophe when simultaneously exposed to DNA damage. Currently, several CHK1 inhibitors (e.g., Sch 900776, GDC0425, GDC0575) are evaluated in clinical trials, highlighting the therapeutic advantage and benefits of a checkpoint abrogation approach in anticancer treatment (48–50).

Interestingly, a recent study showed that CHK1 inhibition potentiated the effect of chemotherapy in NSCLC-derived cancer stem cells and prevented tumor growth in mouse xenografts independently of p53 status (51). Moreover, Zenvirt and colleagues reported that p53-deficient cells are not susceptible to DNA damage following CHK1 inhibition to a greater extent than p53-proficient cells, but the p53 status rather determines the cell death mechanisms and survival. They have shown that while wild-type p53 cells underwent caspase-dependent apoptosis, p53-depleted counterpart resulted in cell death by mitotic catastrophe (52). This outcome is consistent with our current findings in respect to elevated apoptotic events in p53-proficient GTL-16 shC cells upon EMD-1214063 in combination with irradiation. On the other hand, our p53 knockdown line exhibited a comparable caspase-3 activation as well as significantly higher nuclear fragmentation and polyploidy following the combined therapy, which eventually lead to a reduced viability and enhanced cytotoxicity. The discrepancy in these studies may originate from the different targeted approaches and thus from corresponding mechanisms of action. While selective inhibition of CHK1 results mainly in defective DNA damage response, inactivation of the MET receptor particularly in MET-dependent tumors as used in our work induces growth arrest accompanied by the sensitization to DNA damage possibly via checkpoint abrogation. Yet, the long-term cell survival, clonogenicity, and the appropriate preclinical tumor model for this concept are still to be evaluated. Finally, the exact molecular mechanisms and the direct crosstalk between the MET signaling cascade and the DNA damage response machinery require a further extensive investigation.

In this study, we have further demonstrated the potential application of the selective small-molecule MET inhibitor EMD-1214063 as a mean to abrogate the intra-S and G₂–M checkpoint in MET-dependent tumors and have extended this concept by showing its radiosensitizing potential especially in cells with compromised p53 activity. Our results suggest that this targeted approach may overcome the resistance mechanism and its associated poor clinical outcome that originates from deregulated p53 particularly in case of MET-driven tumors. As MET and p53 are aberrantly expressed or mutated, respectively, in numerous human malignant tumors and at significantly high frequencies, the concept of a possible synthetic lethality presented here as a mean of combining p53 deficiency and MET inhibition, thus depleting the cells of two important cell-cycle checkpoints (G₁ and G₂–M, respectively), may be of clinical relevance.

A. Blaukat and F. Bladt are employees of Merck KGaA. No potential conflicts of interest were disclosed by the other authors.

Conception and design: Y. Zimmer, M. Medová, K. Mikami

Development of methodology: M. Medová, K. Mikami, M.P. Tschan

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): K. Mikami, M. Medová, P. Francica, A. A. Glück, L. Nisa

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): K. Mikami, M. Medová, Y. Zimmer, L. Nisa

Writing, review, and/or revision of the manuscript: K. Mikami, M. Medová, Y. Zimmer, D.M. Aebersold, A. Blaukat, F. Bladt, M.P. Tschan

Study supervision: Y. Zimmer, M. Medová

The authors thank Dr. W. Blank-Liss and B. Streit for their excellent technical support.

This study has been supported by Ruth & Arthur Scherbarth Stiftung grant (to M. Medová) and by the Swiss National Science Foundation (grant. no. 31003A_125394), Stiftung zur Krebsbekämpfung and Novartis Foundation (all to Y. Zimmer).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.