lable at ScienceDirect

Biochimie 142 (2017) 135e143

Contents lists avai

Biochimie

journal homepage: www.elsevier .com/locate/biochi

Review

Co-targeting c-Met and DNA double-strand breaks (DSBs):Therapeutic strategies in BRCA-mutated gastric carcinomas

Chrysovalantou Mihailidou a, 1, Michalis V. Karamouzis a, *, 1, Dimitrios Schizas b,Athanasios G. Papavassiliou a, **

a Molecular Oncology Unit, Department of Biological Chemistry, Medical School, National and Kapodistrian University of Athens, 11527 Athens, Greeceb First Department of Surgery, Medical School, National and Kapodistrian University of Athens, 11527 Athens, Greece

a r t i c l e i n f o

Article history:Received 31 July 2017Accepted 4 September 2017Available online 7 September 2017

Keywords:Gastric cancerHepatocyte growth factor receptorc-METBRCA proteinsHR

* Corresponding author. Department of BiologicalUniversity of Athens, 75, M. Asias Street, 11527 Athen** Corresponding author. Department of BiologicalUniversity of Athens, 75, M. Asias Street, 11527 Athen

E-mail addresses: mkaramouz@med.uoa.gr (M.V.uoa.gr (A.G. Papavassiliou).

1 C. Mihailidou and M.V. Karamouzis contributed e

http://dx.doi.org/10.1016/j.biochi.2017.09.0010300-9084/© 2017 Elsevier B.V. and Société Française

a b s t r a c t

Gastric cancer (GC) is a threatening malignancy characterized by heterogeneity. Current therapies useDNA damaging agents, for example, chemotherapeutic agents and ionizing radiation (IR). However, asignificant portion of GC patients develops therapeutic resistance to DNA damage response (DDR) -inducing agents. An important mechanism is the stimulation of the c-MET RTK, which is a tyrosine kinasereceptor and its ligand hepatocyte growth factor (HGF), which facilitates cell survival by boosting DNAdamage repair pathways and via escaping cell cycle arrest. A small subgroup of GC diagnosed patientshas defects in BRCA1 and BRCA2 as mediators of DNA repair proteins. BRCA1/2 related-tumors acquireresistance to chemotherapy through the DSBs (DNA double strand breaks) repair pathways. However,BRCA2-deficient cells, are vulnerable to PARP [poly (ADP-ribose) polymerase] inhibitors as the replicationforks collapse and the DNA-induced damage is not reversed. Herein, we pose that taking into consid-eration the defective DDR machinery can trigger GC cell sensitization to therapies via inhibition of DNArepair response. Inhibition of DNA damage response axis may designate cancer cells with BRCAness(BRCA-mutant cells) more vulnerable to DNA-damaging mediators, such as c-Met inhibitors.

© 2017 Elsevier B.V. and Société Française de Biochimie et Biologie Moléculaire (SFBBM). All rightsreserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1362. The MET/HGF pathway - an overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

2.1. c-MET/HGF - structure and function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1362.2. The impact of HGF/c-MET axis in GC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1362.3. The development of c-MET axis inhibitors in GC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

3. DNA damage signaling on phenotypes of gastric cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1383.1. DNA damage response (DDR) processes and consequential repair pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1383.2. DSB repair pathways - associated BRCA genes in GC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1383.3. DSB repair pathways - associated resistance of GC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1393.4. DSB repair pathways - related BRCA as a target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

4. HGF/c-MET pathway participates in DDR pathways in BRCA-mutated GC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1404.1. MET RTK system affects DSB repair in HRR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1404.2. MET RTK system could affect DNA repair in BRCA-deficient cancer cells in HRR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

Chemistry, Medical School,s, Greece.Chemistry, Medical School,s, Greece.Karamouzis), papavas@med.

qually to this work.

de Biochimie et Biologie Molécul

aire (SFBBM). All rights reserved.

C. Mihailidou et al. / Biochimie 142 (2017) 135e143136

4.3. DSB repair and c-Met as therapeutic co-targets in BRCA-mutated GCs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1405. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

1. Introduction

Gastric cancer (GC) is a tremendously complex heterogeneousdisease with an expanding amount of “driver” mutations havingbeen documented in almost 40% of diagnosed GC [1]. HGF and itsreceptor c-Met, have a fundamental responsibility in the progressionof GC, while its' expression is related to dismal prospects [2]. Gastrictumors are exhibiting constitutive activation of HER familymembersintercede resistance to MET targeted therapy in gastric carcinomacells [3]. Therapeutic agents targeting HGF/MET pathway, such asrilotumumab and onartuzumab, have been industrialized and eval-uated in advanced GC patients [4e6].

The ACRG (Asian Cancer Research Group) had identified fourmolecular subtypes referred on established genetic characteristicsof GCs: MSI (microsatellite instable), MSS (microsatellite stable)/EMT (epithelial-mesenchymal transition), microsatellite stable/tu-mor protein 53 (TP53)þ, and microsatellite stable/tumor protein 53(TP53)�, subtypes [7]. Also, the TCGA (The Cancer Genome Atlas)network had created a four subtype molecular classification systemfor GC referred on the underlying genetic profiling of each subtype:Epstein-Barr virus (EBV) -positive, genomically stable, chromo-somal instability (CIN), microsatellite-instability (MSI) [8]. Reportshave been stated that approximately 8% of GC cases are associatedwith CIN and MSI, indicating to inadequate DNA mismatch repair[7,8]. Interestingly, tumor cells with abnormalities in DDR appa-ratus, become obsessed tomaintain intact repair pathways throughthe NHEJ (non-homologous end joining) pathway. NHEJ helpscancer cells to sustain their growth and simultaneously representsa resistancemechanism to DNA-damaging cytotoxic treatments, forexample, radiochemotherapy. The up-regulated DNA damagesignaling and DNA repair pathways that cancer cells are addicted,may also symbolize cancer's ‘Achilles’ heel’ [9]. Inhibition of DNArepair pathways might probably promote a tumor-selective anti-cancer effect by inhibiting the DNA damage repair via utilizing thetheory of synthetic lethality. PARP [Poly (ADP-Ribose) polymerase]proteins bind to DNA breaks and recruit DNA repair proteins to thelocus of damage. PARP inhibitors promote DSBs, the most harmfulstructure of DNAdamage, exhibiting defective DNA repair mediatedby homologous recombination (HR) repair [10]. The BRCA1-andBRCA2-genes that generate tumor suppressor proteins are crucial tominimize genetic alterations and instability for HR [11,12]. Inheri-ted mutations of BRCA genes are linked to high risk for GC [11].BRCA defective tumors have a tendency to be sensitive to PARPinhibitors [11]. The interaction between PARP and BRCA is animportant synthetic lethal method that could be utilized in thissubset of GCs [1].

Accumulating evidence suggests that MET signaling is linked toDSBs damage response pathways [13]. Depended on these data, wemay identify aberrant MET function conferring acquired resistanceto DNA damaging agents (DDAs). This paper demonstrates a syn-opsis of the c-MET signaling, including its' task in the progression ofGC, and at the same time provides a foundational rationale fortargeting this pathway. Furthermore, we discuss the implications ofthe so far presented clinical data regarding targeting DSB repairpathways in the management of GC. Additionally, we highlight thesignificance of linking c-Met and DSBs repair inhibitors with the

intention of achieving synergistic therapeutic effects in certain GCpatients with BRCAness.

2. The MET/HGF pathway - an overview

2.1. c-MET/HGF - structure and function

c-METwas initially discovered by the early '80s, in an establishedosteogenic sarcoma cell line treatedwith the carcinogen compoundMNNG (N-methyl-N0-nitronitrosoguanidine), by a chromosomal re-organization that combines two genetic loci, the sequence from thetranslocated promoter region (TRP) locus on chromosome 1 to asequence from MET on chromosome 7 with the TRP locus onchromosome 1. Subsequent studies showed that the encoded pro-tein was a receptor tyrosine kinase (RTK). c-MET transcription isinduced by HGF and multiple growth factors, producing an onco-genic protein, the c-MET tyrosine kinase [14,15] [Fig. 1A]. Twodifferent experimental approaches characterized the ligand for c-MET; as a mitogen (stimulation of cell growth) factor for hepato-cytes and as amotility factor for epithelial cells, while this factorwasafterward revealed to be HGF, otherwise called scatter factor (SF)[15,16].

Binding of HGF/SF to its cognate receptor c-MET undergoestyrosine residue phosphorylation and homodimerization, allowingthe recruitment of multiple adaptor proteins and triggeringdownstream stimulation of the phosphoinositide 3-kinase (PI3K)/AKT, RAS/mitogen-activated protein kinase (MAPK), activation oftranscription proteins (STAT), and nuclear factor-kB [17] [Fig. 1B]. Inhepatocytes and placental trophoblast cells, c- HGF and MET offercrucial signals for cell growth, cell survival and cell proliferation forthe period of development. Accordingly, Hgf or MET knockout em-bryos died due to severe developmental defects in the placenta andliver [14]. Nevertheless, abnormalities of the c-MET pathway indi-cate a central role in cell growth, cell proliferation, apoptosis,metastasis and angiogenesis [17,18] [Fig. 1B].

2.2. The impact of HGF/c-MET axis in GC

The significance of c-MET in GC was first identified in GTL-16gastric carcinoma cells with well-documented c-MET gene [19].TPR-MET RNA overexpression was identified in precancerous GClesions [20]. In retrospective studies, elevated expression of c-METhas been reported in about 40% of detecting GC cases and ampli-fication of c-MET gene was reported in 12% of patients. c-Metexpression in tumors is associated with the stage of invasion andmetastasis and dismal prognosis. Moreover, Hs746T, MKN45,NUGC4 and SNU5 gastric cancer cell lines, have MET amplificationand were used in preclinical reports of MET inhibition [21,22]. Re-ports have been also indicated that SOBP-MET (T07) and LACE1-MET (T20) genes are amplified in GC identified patients [23]. InGC, c-METgenemutations are rare. The first one reportedwas in theMET juxtamembrane domain (P1009S) in primary GC [24]. More-over, after genetic analysis of the cytoplasmic domains of c-MET inHs746T gastric cancer cell line an exon 14 mutation of c-MET wasdetected, triggering deletion of the juxtamembrane domain [25].Therefore, accumulated data have demonstrated a key function of

Fig. 1. Hepatocyte growth factor (HGF) and c-MET domain structure, HGF/c-Met signaling and targeted therapy in GC. (A) Scatter factor/Hepatocyte growth factor/(SF/HGF) isproduced primarily by MSCs (mesenchymal stem cells) as an inactive polypeptide (pro-HGF) and is cleaved to its active form by a number of extracellular proteases at specific sitesamong Arg494 and Val495 amino acids. The active form of HGF entails of two polypeptide chains (a- and b-chain), linked by a disulphide bond. The a-chain consist of an N-terminalresidues (hairpin-loop region) followed by Kringle (K) regions. The b-chain consist of serine proteases characterized by deficiency at enzymatic activity. c-Met is composed of adisulphide-linked a/b heterodimer, generated by endoproteolytic cleavage of a common precursor processing in the post-Golgi endomembrane. The extracellular domain of c-Metrepresents three functional domains. The extracellular N-terminal region generates the semaphoring (Sema) region, which includes the whole a-subunit and part of the b-subunit.The PSI domain that spans about 50 residues and contains four disulphide bonds, is linked to the transmembrane helix through four immunoglobulineplexinetranscription (IPT)domains. Intracellularly, the c-Met receptor encompasses Y1234 and Y1235 catalytic tyrosines, which moderate enzyme action, whereas the juxtamembrane tyrosine 1003 is anegative regulatory site, by employing the ubiquitin ligase casitas B-lineage lymphoma (c-CBL). (B) HGF binding to c-Met triggers receptor autophosphorylation and of tyrosineresidues positioned inside the catalytic loop of the tyrosine kinase domain. Subsequent phosphorylation of the C-terminal docking sites (tyrosines 1349 and 1356) of Met allowsphosphorylation and recruitment of signaling effectors that include the adapter proteins GRB2, SHC, Gab1, and Crk/CRKL along with several other signaling proteins. PTPs regulatedephosphorylation of the tyrosines either in the kinase or the docking site via c-Met signaling. Lastly, PLCg binding to c-Met triggers the stimulation of protein kinase C (PKC), whichthen phosphorylates a negatively modulation of c-Met receptor action. Activation of c-Met pathway controls cell cycle progression, cellular proliferation, survival, migration/motility, cell invasion, tumor metastasis and DNA repair, all of which may be enhanced in GCs. Consequently, targeting the HGF/Met signaling pathway with HGF inhibitors(Rilotumumab), c-MET antagonists (Onartuzumab, Foretinib, Tivantinib) and MET kinase inhibitors (Crizotinib, Cabozantinib, AMG-337) may be proved versatile candidates fortargeted therapeutic intervention. (This figure is modified from Organ et al. 2011) [14].

C. Mihailidou et al. / Biochimie 142 (2017) 135e143 137

c-MET and a reasoning for the development of c-MET inhibitors inGC.

2.3. The development of c-MET axis inhibitors in GC

Several agents targeting c-MET have been categorized, devel-oped and clinically tested in GC patients [Fig. 1B, Table 1]. Mono-clonal antibody rilotumumab showed clinical benefit in conditionsof survival and response in randomized phase I/II clinical trials.Moreover, rilotumumab plus chemotherapy resulted in a prolongedprogression-free survival (PFS) when assessed with chemotherapyonly, particularly in diagnosing patients characterized by

overexpression of c-MET [5]. Moreover, rilotumumab was found toalleviate worries about possible cytotoxicities of c-MET-targetedtherapy in healthy tissues [26,27].

In the same vein, a phase III study examining chemotherapy inthe presence or absence onartuzumab in patients with metastaticErbB2-negative and c-MET-positive GC has been terminated due toefficacy problems in the onartuzumab arm [28]. Also, onartuzumabtreatment to first-line mFOLFOX6 did not significantly improveclinical benefits in patients with MET-positive gastroesophagealadenocarcinoma [29]. Additionally, no clear anti-cancer activitywas found by the use of selective small molecule c-MET RTK in-hibitors, excluded of tumors overexpressing c-MET [30]. A

Table 1Current clinical studies of HGF/c-Met inhibitors in GC.

Compound Type Phase ClinicalTrials.gov Identifier Combination

Rilotumumab HGF mAb Phase III NCT02137343 CXPhase III NCT01697072 ECXPhase I/Ib NCT01791374 e

Phase Ib/II NCT00719550 ECXPhase II NCT01443065 FOLFOX

Crizotinib c-MET kinase inhibitor Phase II NCT02435108 e

Phase II NCT02034981 e

AMG-337 c-MET kinase inhibitor Phase I/II NCT02344810 mFOLFOX6Cabozantinib c-MET kinase inhibitor Phase II NCT01755195 e

Phase II NCT00940225 e

Onartuzumab c-MET mAb Phase III NCT01662869 mFOLFOX6Foretinib c-MET mAb Phase II NCT00725712 e

Tivantinib c-MET mAb Phase I/II NCT01611857 FOLFOX

CX:Cisplatin and Capecitabine; ECX:epirubicin; cisplatin & capecitabine; FOLFOX: folinic acid (leucovorin), fluorouracil, oxaliplatin; mFOLFOX6:Oxaliplatin, leucovorin cal-cium, fluorouracil (5FU).

C. Mihailidou et al. / Biochimie 142 (2017) 135e143138

randomized phase II study showed that foretinib, an oral multi-kinase inhibitor targeting c-Met, AXL, TIE-2, RON and VEGFR2 re-ceptors, can decrease phospho-Met (pMet)/total Met tumor ratio[31]. However, these findings showed that foretinib lack effective-ness in a non-selected population of patients with advanced GC[31,32]. In a randomized phase II trial the action of a non-ATP small-molecule c-Met inhibitor Tivantinib displayed modest efficacy inadvanced GC patients [33]. Crizotinib is also a potent small mole-cule c-Met inhibitor modulating c-Met, AKT and ERK signaling axis.Crizotinib has shown promising antineoplastic actions in c-Met-amplified GC patients [34]. Application with AMG 337 influencedthe survival of SNU5 and Hs746T GC cell lines and demonstrated adose-dependent antitumor effectiveness in GC xenograft modelswith c-Met gene amplification [22]. Additional tyrosine kinase in-hibitors, cabozantinib and golvatinib, which block the autophos-phorylation of c-Met are being tested [35] [Table 1].

Finally, recent in vitro and in vivo studies investigated selectivec-Met inhibitors. Combination treatment with PF-04217903; a c-Met inhibitor and PF-04880594; an Raf inhibitor, inhibited phos-phorylation of ERK and reduced cell growth, suggesting that thecombined application of a c-Met tyrosine kinase inhibitor with aMEK inhibitor or a B-Raf inhibitor might be more successful intreatment-resistant tumors that utilize stimulated B-Raf to evadethe inhibition of c-Met signaling in GTL16 cells [36]. Also, a small-molecule kinase inhibitor, T-1840383, that targets VEGFRs and c-Met by inhibiting HGF-stimulated c-Met phosphorylation andactivation of VEGFR has illustrated anticancer action in a peritonealdissemination gastric tumor model [37]. Yhhu3813 [38] and KRC-00715 [39] inhibitors can also reduce c-Met activity and thedownstream signaling pathways, suggesting a potential therapeu-tic activity in c-Met overexpressing GCs. Strong data establishesthat HGF/c-MET inhibitors might need, not only an evaluation ofHGF/c-MET positivity in tumors but also a deeper understanding ofhow and to what level gastric carcinomas are HGF/c-MET linkedwith ligand-dependent/or independent signaling axis [22].

3. DNA damage signaling on phenotypes of gastric cancer

3.1. DNA damage response (DDR) processes and consequentialrepair pathways

Once the DDR machinery detects DNA damage lesions, it re-cruits a complicated and entangled network to guarantee that thetranscription and translation actions are paused by preventingprogression from one stage of the cell cycle to the next stage, eitherallowing time to the cell to fix the damage or promote cell apoptosis[40,41]. The major mediators in DDR pathways are members of the

protein kinase PI3K (phosphatidylinositol 3-kinase) family,including Rad 3-related (ATR), ataxia-telangiectasia mutated (ATM)and DNA-dependent protein kinase (DNA-PK). When DNA damagelesions are detected by a sensor protein, these mediators arerecruited to the site of the damage and simultaneously phosphor-ylation of downstream proteins, that are involved in the DDR, isoccurred [40]. In addition to these mediators, poly (ADP-ribose)polymerases (PARPs), are well-documented elements with impor-tant roles in DDR [42]. PARP-1 and PARP-2 have been proposed toplay a crucial role in single-strand DNA-binding protein [ssDNA-binding protein (SSB)] by its' rapid binding to SSB and subsequentrounds of SSB detection and signaling [43e45]. PARP-3 has newlybeen acknowledged to also have a task in facilitating NHEJ [46]. Thedevelopment of several PARP inhibitors (PARPi) is considered tohave a promising role in cancer therapeutic strategy, as they areappearing to cause the most lethal structure of DNA damage, theDSBs [43]. The failure of DSB repair can promote chromosomalaberrations (genomic instability) and increased tumorigenesis.Homologous recombination (HR) and nonhomologous end joining(NHEJ) are themost crucial DSB repair pathways in yeast and highereukaryotes [47,48]. DSBs are repaired by NHEJ in the G1 phase,whereas HR occurs during the end of S and G2 phases of the cellcycle [Fig. 2].

3.2. DSB repair pathways - associated BRCA genes in GC

BRCA1 and BRCA2 are breast cancer susceptibility proteins, bothmandatory components of HR, although they have different roles inthe assembly of HR. BRCA1 participates in the early embryonicdevelopment, interacts with proteins that have a key role in cellcycle progression (cell cycle checkpoints) during the end of S andG2 phases. As a consequence BRCA1 promotes apoptosis inresponse to DNA lesions and initiates HR. The main function ofBRCA2 is to interact with BRCA1 upon DNA damage [49,50]. BRCA1or BRCA2 gene mutations might result in HR silencing. BRCA defi-ciency leads to genome translocations and aberrant rearrange-ments, that can develop cancer in normal healthy tissues [51].Multiple studies suggest a connection between BRCA1/2 mutationsand GC [1,30]. The mutation of the BRCA2 gene, rather than BRCA1gene, has a greater impact on the risk of developing GC as well as inthe survival of patients. BRCA1/2 abnormalities have been demon-strated to increase the risk of developing gastric cancer by as muchas six fold increase among first-degree relatives of BRCA1/2 ab-normalities carriers. The risk is reported to be fourfold higher inBRCA1 abnormalities carriers and at least twofold higher in BRCA2abnormalities carriers [11]. However, it has also been proposed thatwhile there are apparent correlations among ovarian, prostate,

Fig. 2. Homologous Recombination and Nonhomologous DNA End Joining-Mediated DNA Double-Strand Break Repair mechanisms in humans. After recruitment of the DSB byMre11-Rad50-Nbs1 (MRN) protein complex, ATM is activated and H2AX is phosphorylated, which sequentially recognizes a sequence of signaling actions that results in thestimulation of nucleases, such as CtIP and Mre11 to create ends with ssDNA “3'overhangs”. RPA binds strongly to ssDNA, which is next replaced by Rad51 and Rad51 paralogs thatare helped by Rad52, Rad54 and BRCA2. The following addition of RPA to incompletely formed Rad51-ssDNA filaments, starts strand invasion into an undamaged homologous DNAmolecule, promoting the formation of a Holiday junction. The DNA sequence around the DSB is copied by DNA synthesis related to branch migration and the process is accom-plished by cleavage of the Holliday junction. HRR is an error-free process as a consequence of the existence of homologous sister chromatids. NHEJ can be separated into twopathways; C-NHEJ (canonical NHEJ) and A-NHEJ (alternative NHEJ), according to the resection of DNA ends at the breakage site and the proteins involved. For C-NHEJ, Ku proteinsbind to the broken ends of DSBs, which recruits PNKP or Tdp1 and DNA-PKcs. Upon end-binding, DNA-PKcs is stimulated, auto-phosphorylated and phosphorylates H2AX.Phosphorylated DNA-PKcs permits the DNA ligase IV/XRCC4/XLF complex to link the broken DNA ends, possibly relied on a DNA polymerase to fill gaps. Back up pathway is thealternative (A-NHEJ) pathway. There is additional evidence that higher eukaryotic cells use the PARP-1/DNA Ligase III/XRCC1 repair module and that its' function is expedited via thelinker histone H1. Recent genetic studies identified that polymerase theta (Polq) is capable to extend the 30 ssDNA mediating a combine of template-dependent/or independentactivities. (This figure is modified from Iliakis et al. 2015) [40].

C. Mihailidou et al. / Biochimie 142 (2017) 135e143 139

breast, pancreatic and gastric malignancies, it is likely that suchassociations may be caused by additional factors than having aBRCA1 or BRCA2 gene mutation [10,52].

3.3. DSB repair pathways - associated resistance of GC

The NHEJ mechanism is regulated by the DNA-PK complexconsisting of DNA-PKCS, a catalytic subunit, and its heterodimer Kuregulatory subunits [p86 (Ku 80) and p70 (Ku 70) [Fig. 2]. The DNA-PK complex attaches to DNA ends and functions as a scaffold foradditional DNA repair proteins. Activation of DSB repair enzymescontributes to chemotherapy and radiotherapy resistance. Proteinlevels of KU and DNAPKcs have been reported to be increased in GCcell lines [53]. Additionally, the level of gH2AX foci, a sensor of DSBrepair, remain elevated in gastric cancer cell line N87, treated withradiation [53]. Higher levels of KU have been established withelevated DNA binding potential and increasing radiosensitization incancer cells, demonstrating advanced repair [54]. Additionally, ithas been documented that over-activation of DSB repair pathways

is associated with chemoresistance in GC [55].

3.4. DSB repair pathways - related BRCA as a target

Mutations in X-ray cross complementing group 1 (XRCC1) andin BRCA1 seem to have a positive impact on taxane and cisplatintherapy in GC patients [56]. XRCC1 takes part in the single nucle-otide excision repair and it is also a player in the process of the A-NHEJ [57]. It has been also found that in BRCA1 positive carcinomas,the protein XRCC1 has been often deregulated, suggesting that theA-NHEJ pathway may be stimulated in these cells [56]. Moreover,PARP1 gene expression and activation are found in deficient cancercell lines, which is another hint that components of the A-NHEJpathway are overactive in BRCA associated tumors [57]. A recentstudy has shown that BRCA1-deficient cancers were highly sensi-tive to the inhibition of eNHEJ [58]. It was also demonstrated thatpolymerase theta (Polq) suppresses alternative NHEJ, suggestingthat abolishment of Polq has a synergistic outcome on tumor cellsurvival lacking BRCA genes [58]. BRCA1/PARP combination

C. Mihailidou et al. / Biochimie 142 (2017) 135e143140

treatment has raised as a potential novel treatment approach in theera of precision cancer therapy [59]. PARP inhibitors have beenevaluated in many carcinomas with promising results [60,61]. Thesuccess of their use in patients with ovarian cancer has generatedgreat expectations [62]. Although the occurrence of BRCA muta-tions in GC is small, it has been found that the employ of PARPinhibitors in GC cell lines has additive efficacy combined withchemotherapeutic drugs [63]. A Phase II study (NCT01063517) re-ported that the olaparib, a PARP inhibitor, when combined withpaclitaxel showed a statistically significant survival benefitcompared with paclitaxel alone in Asian patients whose tumorshad a lower protein expression for ATM with advanced gastriccarcinoma. A subsequent Phase III study (GOLD, NCT01924533)assessed the efficacy of olaparib in combination with paclitaxelversus placebo with paclitaxel. It was observed that the overallsurvival was independent of ATM protein expression status [64].Based on these findings, clinical studies are examining the appli-cation of PARP inhibitors in patients diagnosed with GC and tryingto identify reliable predictive molecular factors based on the betterunderstanding of DSB repair pathways [65e67].

4. HGF/c-MET pathway participates in DDR pathways in BRCA-mutated GC

4.1. MET RTK system affects DSB repair in HRR

c-Met holds a crucial role in DDR and especially in signalingpathways which are implicated in DNA repair. HR DSBs repair ef-fectors such as c-Met and RAD51 recombinase and V-abl Abelsonmurine leukemia viral oncogene homolog 1 (ABL) have been sug-gested [13]. Cells expressing c-Met-mutated variants were treatedwith c-Met inhibitors. This was accompanied with reduced RAD51phosphorylation, impairing its' nuclear translocation upon DNAdamage [13]. These events cause diminished capacity for DNArepair, as confirmed by the capability of SU11274 to postponedecrease in gH2AX levels, that negatively affects HR-related cellularcomponents [13]. Yu et al. reported that SU11274, a c-Met inhibitor,enhanced the radiosensitivity of DU145 human prostate cancer celllines mediated by failure of DNA repair capability, abolishment ofcell cycle arrest and augmentation of cell death. This research wasamong the first to prove the success of combining c-Met inhibitorstogether with radiotherapy in cancer therapeutics [68].

Another study has shown that c-Met inhibition in cancer cellsaffects the ability to properly execute DSB repair and notably de-teriorates HR in a dose-dependent approach, confirming that theobserved HR inhibition is c-Met-dependent. More mechanistic in-sights showed that c-Met inhibition influences the pattern of theRAD51-BRCA2 complex, which is critical for error-free HR repair ofdouble strand DNA damage, apparently upon Rad51 down-regulation and impaired translocation into the nucleus [69]. Asynergistic model of interaction between c-Met inhibition com-bined with ionizing radiation was adequate, to augment gH2AXlevels in GTL-16, human gastric adenocarcinoma cells, stronglyrepresenting the accumulation of double-strand DNA breaks [70].Finally, c-Met inhibition diminishes the stimulation of CHK1, ATRand CDC25B in GTL-16 cells as well as nullify an associated DNAdamagee intra-S-phase cell-cycle arrest. These data suggest that c-Met inhibition represents a significant damage-checkpoint thatmight facilitate DNA-defective cells to escape cell cycle arrestbefore a repair is accomplished [71].

During cellular DSB response, chromatin undergoes rearrange-ment discriminated by H2A.X Ser139 phosphorylation (g-H2A.X).gH2AX serves as a component between the equilibrium of DNArepair and apoptosis response, through a post-translational change,phosphorylation of tyrosine 142, being able to affect the

recruitment to gH2AX of functional repair/survival complexes.Combinations of c-Met-targeted therapy and irradiation, increasedgH2AX tyrosine phosphorylation levels and pro-apoptotic kinaseJNK1 levels, providing an additional link to c-Met inhibitors andDDR machinery [72,73]. Liu et al. have revealed that such combi-nations can suppress HGF-induced cell proliferation and phos-phorylation of the ERK1/2, AKT, and STAT3 [74], enhance theformation of DNA double strand breaks and probably alleviate tu-mor hypoxia [75]. Bacco and his research team have confirmed thatradiation influence the activity and expression levels of c-Met viathe ATM-NF-kB signaling pathway resulted in cell invasion.Therefore, c-Met inhibition increase tumor cell radiosensitivity[76]. These results indicate the function of c-Met in the DDRapparatus and the causality of its' targeting, in order to increasecancer cells sensitivity to DNA-damaging mediators.

4.2. MET RTK system could affect DNA repair in BRCA-deficientcancer cells in HRR

PARP inhibitors are successfully eliminated tumors defective inBRCA/2 genes due to the model of synthetic lethality [77e79].Earlier studies have revealed that PARP inhibitors activate g-H2AXand RAD51 foci formation. Cells lacking PARP1, induce homologousrecombination by collapsed replication forks [80,81]. Additionally,BRCA2-deficient cells, as a consequence of their deficiency in ho-mologous recombination, are intensely susceptible to PARP in-hibitors, resulting in DSBs in S-phase and no longer repaired [80].This renders BRCA-deficient cells susceptible to agents, for instance,PARP inhibitors, that are provisionally ‘synthetic lethal' throughtheir principal repair defect [82e84]. Developing drugs, such as aninhibitor of DNA repair that constantly kill cancer cells, in the lackof an exogenous DNA-damaging factor, illustrates a novel model.

Homologous recombination (HR) also plays a critical role in themaintenance of genome integrity. RAD52 is an essential member ofthe HR pathway [83]. c-Met activation affects Rad51 regulation andtranslocation into the nucleus [69]. Combinations of RAD52 muta-tions with mutations in genes like RAD51C BRCA1/2, PALB2 are le-thal [83]. Accordingly, small molecule inhibitors of RAD52 mightblock the biological and chemical functions of RAD52 and theendless growth of BRCA1-deficient cells, representing a potentiallyimportant target for GC therapy [85]. The role of c-MET in the othertwo DSB repair pathways C-NHEJ and alternative NHEJ remainsuncertain at the moment. Further research will be needed tosensitize tumor cells to currently DDR.

4.3. DSB repair and c-Met as therapeutic co-targets in BRCA-mutated GCs

Deregulated HGF/MET signaling is observed in an extensiverange of malignancies, including GC [2]. While the current findingsreport that signaling via the cMET protects tumor cells from DNAdamage [70], cellular events closely linking MET to the DDR ma-chinery remain currently unknown. New insights into these bio-logical networks evolve as an emerging necessity.

PARP inhibitors are licensed by the FDA for cancer treatmentwith BRCA mutation carriers. Importantly, recent studies showevidence that BRCA1-deficient cells may develop resistance duringthis treatment, partly via correcting their defect in HR-mediatedDSB repair, or via synthetic viable loss of 53BP1 or through ge-netic reversionmutations in BRCA1. In addition, phosphorylation ofPARP1 at Tyr907 increases PARP1 enzymatic activity and eliminatesbinding to a PARP inhibitor, capable of developing cancer resistantto PARP inhibitors [61,62,81]. Du et al. have examined the effects ofthe dual inhibition of c-Met and PARP, against cancer. Both com-binations exhibited a synergistic inhibitory effect on clonogenic cell

Fig. 3. Combined inhibition of both c-Met and DNA repair mediators may exhibit synergistic therapeutic effects in GC. c-Met deregulation and resultant aberrant of HGF/c-Metsignaling may happen via several different mechanisms. The HGF/c-Met pathway has been highlighted into an increasingly attractive key target for antitumor therapeutics.Recently, various molecules targeting HGF/c-Met have been assessed in clinical trials, involving monoclonal antibodies and inhibitors either ligand-mediated targeting or thereceptor-mediated targeting. Recent data indicate that c-Met inhibitors induce apoptosis, possibly through increasing/enhancing DNA-induced damage and alteration of DNAdamage/repair proteins. However, tumor cells can continue proliferating by over-activating a variety of DNA damage response and DNA repair pathways to abolish the causeddamage. In this setting, it is tempting to note that DNA repair mechanisms are designed to be a fundamental target to enhance the efficacy of cancer therapeutics and abolish theDNA damaging agents resistance. The synergistically combined treatment of c-Met and DNA repair inhibitors represents a promising strategy for c-Met-expressing GCs withBRCAness.

C. Mihailidou et al. / Biochimie 142 (2017) 135e143 141

survival in MDA-MB-231, BT549 and HCC1937 TNBC cells. Notably,combination therapy in xenograft tumor models using MDA-MB-231, significantly abrogated tumor growth, increased apoptosis,reduced cell proliferation and superior DNA damage (g-H2AXmarker) was observed, compared to either inhibitor alone. Also,synergistic cell growth inhibitionwas assessed in MCF-7/c-Met andc-Met-expressing NSCLC (non-small cell lung cancer) xenografttumor models. These findings set the hypothesis that this combi-nation therapy may be significantly more effective than PARP in-hibition alone, regardless of the cancer stage and type [81].

Francica et al. [86] reported that MET inhibitors, in MET-overexpressing gastric tumors, trigger the ability of cancer cells tofix the damaged DNA and to enhance the effectiveness of the un-dergoing radiation therapy. Thus, it seems rational to assesswhether the synergistic inhibition of DNA DSB repair via HR orNHEJ and c-Met simultaneously, reveals an efficient therapy inBRCA-mutated gastric carcinomas [Fig. 3].

5. Conclusion

c-MET/HGF pathway has become a tempting candidate againstGC. In addition to its own unique effects, c-Met promotes cellmigration and proliferation by boosting DNA damage repair

pathways and escaping cell cycle arrest. Numerous studies haveconfirmed that a subgroup of GC patients develops BRCA deficiency.Recently, research indicates that BRCA-deficient tumors are morevulnerable to DNA damage since replicating damaged DNA increasethe risk of cell death. We, therefore, propose that co-inhibition ofHGF/c-MET with DNA damage response proteins may represent ahopeful therapeutic approach for c-Met-expressing gastric carci-nomas with BRCAness.

Conflict of interest

No potential conflicts of interest were disclosed.

References

[1] L.B. Alexandrov, S. Nik-Zainal, H.C. Siu, S.Y. Leung, M.R. Stratton, A mutationalsignature in gastric cancer suggests therapeutic strategies, Nat. Commun. 6(2015) 8683.

[2] J. Zhang, L. Guo, X. Liu, W. Li, J. Ying, MET overexpression, gene amplificationand relevant clinicopathological features in gastric adenocarcinoma, Onco-target 8 (2017) 10264e10273.

[3] N. Takahashi, K. Furuta, H. Taniguchi, Y. Sasaki, H. Shoji, Y. Honma, et al.,Serum level of hepatocyte growth factor is a novel marker of predicting theoutcome and resistance to the treatment with trastuzumab in HER2-positivepatients with metastatic gastric cancer, Oncotarget 7 (2016) 4925e4938.

C. Mihailidou et al. / Biochimie 142 (2017) 135e143142

[4] T. Iveson, R.C. Donehower, I. Davidenko, S. Tjulandin, A. Deptala, M. Harrison,Rilotumumab in combination with epirubicin, cisplatin, and capecitabine asfirst-line treatment for gastric or oesophagogastric junction adenocarcinoma:an open-label, dose de-escalation phase 1b study and a double-blind, rand-omised phase 2 study, Lancet Oncol. 15 (2014) 1007e1018.

[5] S. Doshi, P.O. Gisleskog, Y. Zhang, M. Zhu, K.S. Oliner, E. Loh, et al., Rilotu-mumab exposure-response relationship in patients with advanced or meta-static gastric cancer, Clin. Cancer Res. 21 (2015) 2453e2461.

[6] M.A. Shah, J.-Y. Cho, I.B. Tan, N.C. Tebbutt, C.-J. Yen, A. Kang, et al.,A randomized phase II study of FOLFOX with or without the MET inhibitoronartuzumab in advanced adenocarcinoma of the stomach and gastroesoph-ageal junction, The Oncologist 21 (2016) 1085e1090.

[7] R. Cristescu, J. Lee, M. Nebozhyn, K.M. Kim, J.C. Ting, S.S. Wong, et al., Mo-lecular analysis of gastric cancer identifies subtypes associated with distinctclinical outcomes, Nature 21 (2015) 449e456.

[8] T. Chen, X.Y. Xu, P.H. Zhou, Emerging molecular classifications and therapeuticimplications for gastric cancer, Chin. J. Cancer 35 (2016) 49.

[9] A.J. Davis, D.J. Chen, DNA double strand break repair via non-homologous end-joining, Transl. Cancer Res. 2 (2013) 130e143.

[10] S.H. Park, K.Y. Jang, M.J. Kim, S. Yoon, Y. Jo, S.M. Kwon, et al., Tumor sup-pressive effect of PARP1 and FOXO3A in gastric cancers and its clinical im-plications, Oncotarget 6 (2015) 44819e44831.

[11] H. Cavanagh, K.M.A. Rogers, The role of BRCA1 and BRCA2 mutations inprostate, pancreatic and stomach cancers, Hered. Cancer Clin. Pract. 13 (2015)16.

[12] T. Helleday, The underlying mechanism for the PARP and BRCA syntheticlethality: clearing up the misunderstandings, Mol. Oncol. 5 (2011) 387e393.

[13] M. Medov�a, D.M. Aebersold, Y. Zimmer, The molecular crosstalk between theMET receptor tyrosine kinase and the DNA damage response-biological andclinical aspects, Cancers 6 (2014) 1e27.

[14] S.L. Organ, M.-S. Tsao, An overview of the c-MET signaling pathway, Ther. Adv.Med. Oncol. 3 (2011) S7eS19.

[15] L. Teng, J. Lu, CMET as a potential therapeutic target in gastric cancer (Review),Int. J. Mol. Med. 32 (2013) 1247e1254.

[16] M.V. Karamouzis, P.A. Konstantinopoulos, A.G. Papavassiliou, Targeting METas a strategy to overcome crosstalk-related resistance to EGFR inhibitors,Lancet Oncol. 10 (2009) 709e717.

[17] S. Peters, A.A. Adjei, MET: a promising anticancer therapeutic target, Nat. Rev.Clin. Oncol. 9 (2012) 314e326.

[18] H. Kawakami, I. Okamoto, T. Arao, W. Okamoto, K. Matsumoto, H. Taniguchi,et al., MET amplification as a potential therapeutic target in gastric cancer,Oncotarget 4 (2013) 9e17.

[19] C.R. Maroun, T. Rowlands, The Met receptor tyrosine kinase: a key player inoncogenesis and drug resistance, Pharmacol. Ther. 142 (2014) 316e338.

[20] Y. Sun, M.M. Tian, L.X. Zhou, W.C. You, J.Y. Li, Value of c-Met for predictingprogression of precancerous gastric lesions in rural Chinese population, Chin.J. Cancer Res. 24 (2012) 18e22.

[21] L. Marano, R. Chiari, A. Fabozzi, F. De Vita, V. Boccardi, G. Roviello, et al., c-Mettargeting in advanced gastric cancer: an open challenge, Cancer Lett. 365(2015) 30e36.

[22] M. Inokuchi, S. Otsuki, Y. Fujimori, Y. Sato, M. Nakagawa, K. Kojima, Clinicalsignificance of MET in gastric cancer, World J. Gastrointest. Oncol. 7 (2015)317e327.

[23] S.S. Wong, K.M. Kim, J.C. Ting, K. Yu, J. Fu, S. Liu, et al., Genomic landscape andgenetic heterogeneity in gastric adenocarcinoma revealed by whole-genomesequencing, Nat. Commun. 5 (2014) 5477.

[24] Y. Asaoka, M. Tada, T. Ikenoue, M. Seto, M. Imai, K. Miyabayashi, et al., Gastriccancer cell line Hs746T harbors a splice site mutation of c-Met causing jux-tamembrane domain deletion, Biochem. Biophys. Res. Commun. 394 (2010)1042e1046.

[25] J.H. Tong, S.F. Yeung, A.W.H. Chan, L.Y. Chung, S.L. Chau, R.W.M. Lung, et al.,MET amplification and exon 14 splice site mutation define unique molecularsubgroups of non-small cell lung carcinoma with poor prognosis, Clin. CancerRes. 22 (2016) 3048e3056.

[26] K. Sudo, Y. Yamada, Advancing pharmacological treatment options foradvanced gastric cancer, Expert Opin. Pharmacother. 16 (2015) 2293e2305.

[27] D. Cunningham, S.-E. Al-Batran, I. Davidenko, D.H. Ilson, A.M. Murad,N.C. Tebbutt, et al., RILOMET-1: an international phase III multicenter, ran-domized, double-blind, placebo-controlled trial of rilotumumab plus epi-rubicin, cisplatin, and capecitabine (ECX) as first-line therapy in patients withadvanced MET-positive gastric or gastroesophageal junction (G/GEJ) adeno-carcinoma, J. Clin. Oncol. 31 (2013). TPS4153eTPS4153.

[28] M.A. Shah, Y.-J. Bang, F. Lordick, J. Tabernero, M. Chen, S.P. Hack, et al.,METGastric: a phase III study of onartuzumab plus mFOLFOX6 in patientswith metastatic HER2-negative (HER2-) and MET-positive (METþ) adeno-carcinoma of the stomach or gastroesophageal junction (GEC), J. Clin. Oncol.33 (2015) 4012.

[29] M.A. Shah, Y.-J. Bang, F. Lordick, M. Alsina, M. Chen, S.P. Hack, et al., Effect offluorouracil, leucovorin, and oxaliplatin with or without onartuzumab inHER2-negative, MET-positive gastroesophageal adenocarcinoma, JAMA Oncol.(2016) 1e8.

[30] M.A. Shah, Z.A. Wainberg, D.V.T. Catenacci, H.S. Hochster, J. Ford, P. Kunz, etal., Phase II study evaluating 2 dosing schedules of oral foretinib(GSK1363089), cMET/VEGFR2 inhibitor, in patients with metastatic gastriccancer, PLoS ONE 8 (2013) e54014.

[31] B. Cetin, O. Gumusay, M. Cengiz, A. Ozet, Advances of molecular targetedtherapy in gastric cancer, J. Gastrointest. Cancer 47 (2016) 125e134.

[32] N. Baniak, J.-L. Senger, S. Ahmed, S.C. Kanthan, R. Kanthan, Gastric biomarkers:a global review, World J. Surg. Oncol. 14 (2016) 212.

[33] Y.K. Kang, K. Muro, M.H. Ryu, H. Yasui, T. Nishina, B.Y. Ryoo, et al., A phase IItrial of a selective c-Met inhibitor tivantinib (ARQ 197) monotherapy as asecond- or third-line therapy in the patients with metastatic gastric cancer,Investig. New Drugs 32 (2014) 355e361.

[34] W. Xu, Z. Yang, N. Lu, Molecular targeted therapy for the treatment of gastriccancer, J. Exp. Clin. Cancer Res. CR 35 (2016) 1, http://dx.doi.org/10.1186/s13046-015-0276-9.

[35] I. Riquelme, K. Saavedra, J.A. Espinoza, et al., Molecular classification of gastriccancer: towards a pathway-driven targeted therapy, Oncotarget 6 (2015)24750e24779.

[36] H.Y. Zou, Q. Li, J.H. Lee, M.E. Arango, K. Burgess, M. Qiu, et al., Sensitivity ofselected human tumor models to PF-04217903, a novel selective c-Met kinaseinhibitor, Mol. Cancer Ther. 11 (2012) 1036e1047.

[37] Y. Awazu, K. Nakamura, A. Mizutani, Y. Kakoi, H. Iwata, S. Yamasaki, et al.,A novel inhibitor of c-Met and VEGF receptor tyrosine kinases with a broadspectrum of in vivo antitumor activities, Mol. Cancer Ther. 12 (2013)913e924.

[38] C. He, J. Ai, W. Xing, Y. Chen, H. Zhang, M. Huang, et al., Yhhu3813 is a novelselective inhibitor of c-Met kinase that inhibits c-Met-dependent neoplasticphenotypes of human cancer cells, Acta Pharmacol. Sin. 35 (2014) 89e97.

[39] C.H. Park, S.Y. Cho, J. Du Ha, H. Jung, H.R. Kim, C.O. Lee, et al., Novel c-Metinhibitor suppresses the growth of c-Met-addicted gastric cancer cells, BMCCancer 16 (2016) 35.

[40] G. Iliakis, T. Murmann, A. Soni, Alternative end-joining repair pathways arethe ultimate backup for abrogated classical non-homologous end-joining andhomologous recombination repair: implications for the formation of chro-mosome translocations, Mutat. Res. Genet. Toxicol. Environ. Mutagen. 793(2015) 166e175.

[41] A.G. Pallis, M.V. Karamouzis, DNA repair pathways and their implication incancer treatment, Cancer Metastasis Rev. 29 (2010) 677e685.

[42] C.L. Scott, E.M. Swisher, S.H. Kaufmann, Poly (ADP-ribose) polymerase in-hibitors: recent advances and future development, J. Clin. Oncol. Off. J. Am.Soc. Clin. Oncol. 33 (2015) 1397e1406.

[43] J. Yelamos, J. Farres, L. Llacuna, C. Ampurdanes, J. Martin-Caballero, PARP-1and PARP-2: new players in tumour development, Am. J. Cancer Res. 1 (2011)328e346.

[44] W. Ma, C.J. Halweg, D. Menendez, M.A. Resnick, Differential effects ofpoly(ADP-ribose) polymerase inhibition on DNA break repair in human cellsare revealed with Epstein-Barr virus, Proc. Natl. Acad. Sci. U. S. A. 109 (2012)6590e6595.

[45] S.O. Ali, F.A. Khan, M.A. Galindo-Campos, J. Y�elamos, Understanding specificfunctions of PARP-2: new lessons for cancer therapy, Am. J. Cancer Res. 6(2016) 1842e1863.

[46] G.J. Grundy, S.L. Rulten, R. Arribas-Bosacoma, K. Davidson, Z. Kozik, A. Oliver,et al., The Ku-binding motif is a conserved module for recruitment andstimulation of non-homologous end-joining proteins, Nat. Commun. 7 (2016)1e7.

[47] R. Ceccaldi, B. Rondinelli, A.D. D'Andrea, Repair pathway choices and conse-quences at the double-strand break, Trends Cell Biol. 26 (2016) 52e64.

[48] A. Chakraborty, N. Tapryal, T. Venkova, N. Horikoshi, R.K. Pandita, A.H. Sarker,et al., Classical non-homologous end-joining pathway utilizes nascent RNA forerror-free double-strand break repair of transcribed genes, Nat. Commun. 7(2016) 13049.

[49] J. Saha, A.J. Davis, Unsolved mystery: the role of BRCA1 in DNA end-joining,J. Radiat. Res. (2016) i18ei24.

[50] R. Roy, J. Chun, S.N. Powell, BRCA1 and BRCA2: important differences withcommon interests, Nat. Rev. Cancer 12 (2012), 372e372.

[51] C. Jekimovs, E. Bolderson, A. Suraweera, M. Adams, K.J. O'Byrne, D.J. Richard,Chemotherapeutic compounds targeting the DNA double-strand break repairpathways: the good, the bad, and the promising, Front. Oncol. 4 (2014) 86.

[52] M. Lawniczak, A. Jakubowska, A. Bialek, J. Lubinski, K. Jaworska-Bieniek,K. Kaczmarek, et al., BRCA1 founder mutations do not contribute to increasedrisk of gastric cancer in the Polish population, Hered. Cancer Clin. Pract. 14(2016) 3.

[53] M.T. Niazi, G. Mok, M. Heravi, L. Lee, T. Vuong, R. Aloyz, et al., Effects of DNA-dependent protein kinase inhibition by NU7026 on DNA repair and cell sur-vival in irradiated gastric cancer cell line N87, Curr. Oncol. 21 (2014) 91e96.

[54] T. Aparicio, R. Baer, J. Gautier, DNA double-strand break repair pathway choiceand cancer, DNA Repair 19 (2014) 169e175.

[55] B. Altan, T. Yokobori, M. Ide, T. Bai, T. Yanoma, A. Kimura, et al., Highexpression of MRE-11-RAD50-NB51 is associated with poor prognosis andchemoresistance in gastric cancer, Anticancer Res. 36 (2016) 5237e5247.

[56] H.J. Shim, J.Y. Yun, J.E. Hwang, W.K. Bae, S.H. Cho, J.H. Lee, et al., BRCA1 andXRCC1 polymorphisms associated with survival in advanced gastric cancertreated with taxane and cisplatin, Cancer Sci. 101 (2010) 1247e1254.

[57] S. De Summa, R. Pinto, B. Pilato, D. Sambiasi, L. Porcelli, G. Guida, et al.,Expression of base excision repair key factors and miR17 in familial andsporadic breast cancer, Cell Death Dis. 5 (2014) e1076.

[58] P.A. Mateos-Gomez, F. Gong, N. Nair, K.M. Miller, E. Lazzerini-Denchi, A. Sfeir,Mammalian polymerase q promotes alternative NHEJ and suppressesrecombination, Nature 518 (2015) 254e257.

C. Mihailidou et al. / Biochimie 142 (2017) 135e143 143

[59] A. Cerrato, F. Morra, A. Celetti, Use of poly ADP-ribose polymerase [PARP]inhibitors in cancer cells bearing DDR defects: the rationale for their inclusionin the clinic, J. Exp. Clin. Cancer Res. CR 35 (2016) 179.

[60] U. Malapelle, F. Morra, G. Ilardi, R. Visconti, F. Merolla, A. Cerrato, et al., USP7inhibitors, downregulating CCDC6, sensitize lung neuroendocrine cancer cellsto PARP-inhibitor drugs, Lung Cancer 107 (2017) 41e49.

[61] J.S. Brown, B. O'Carrigan, S.P. Jackson, T.A. Yap, Targeting DNA repair in cancer:beyond PARP inhibitors, Cancer Discov. 7 (2017) 20e37.

[62] A. Papa, D. Caruso, M. Strudel, S. Tomao, F. Tomao, Update on Poly-ADP-ribosepolymerase inhibition for ovarian cancer treatment, J. Transl. Med. 14 (2016)267.

[63] M.R. Mirza, B.J. Monk, J. Herrstedt, A.M. Oza, S. Mahner, A. Redondo, et al.,Niraparib maintenance therapy in platinum-sensitive, recurrent ovariancancer, N. Engl. J. Med. 375 (2016) 2154e2164.

[64] Y.-J. Bang, N. Boku, K. Chin, K.-W. Lee, S.H. Park, S. Qin, S.Y. Rha, L. Shen, N. Xu,S.-A. Im, G. Locker, P. Rowe, X. Shi, D. Hodgson, Y.-Z. Liu, R. Xu, Gastrointes-tinal tumours, non-colorectal Olaparib in combination with paclitaxel in pa-tients with advanced gastric cancer who have progressed following first-linetherapy: phase III GOLD study, Ann. Oncol. 27 (suppl_6) (2016) LBA25.

[65] Y.-J. Bang, S.-A. Im, K.-W. Lee, J.Y. Cho, E.-K. Song, K.H. Lee, et al., Randomized,double-blind phase II trial with prospective classification by ATM protein levelto evaluate the efficacy and tolerability of olaparib plus paclitaxel in patientswith recurrent or metastatic gastric cancer, J. Clin. Oncol. 33 (2015)3858e3865.

[66] E. Fontana, E.C. Smyth, Novel targets in the treatment of advanced gastriccancer: a perspective review, Ther. Adv. Med. Oncol. 8 (2016) 113e125.

[67] V.V. Subhash, S.H. Tan, M.S. Yeo, F.L. Yan, P.C. Peethala, N. Liem, et al., ATMexpression predicts veliparib and irinotecan sensitivity in gastric cancer bymediating P53-Independent regulation of cell cycle and apoptosis, Mol.Cancer Ther. 15 (2016) 3087e3096.

[68] H. Yu, X. Li, S. Sun, X. Gao, D. Zhou, C-Met inhibitor SU11274 enhances theresponse of the prostate cancer cell line DU145 to ionizing radiation, Biochem.Biophys. Res. Commun. 427 (2012) 659e665.

[69] M. Medov�a, D.M. Aebersold, Y. Zimmer, MET inhibition in tumor cells byPHA665752 impairs homologous recombination repair of DNA double strandbreaks, Int. J. Cancer J. Int. Du Cancer 130 (2012) 728e734.

[70] M. Medov�a, D.M. Aebersold, W. Blank-Liss, B. Streit, M. Medo, S. Aebi, et al.,MET inhibition results in DNA breaks and synergistically sensitizes tumor cellsto DNA-damaging agents potentially by breaching a damage-induced check-point arrest, Genes Cancer 1 (2010) 1053e1062.

[71] J.W. Welsh, D. Mahadevan, R. Ellsworth, L. Cooke, D. Bearss, B. Stea, The c-Metreceptor tyrosine kinase inhibitor MP470 radiosensitizes glioblastoma cells,Radiat. Oncol. (London, England) 4 (2009) 69.

[72] P.J. Cook, B.G. Ju, F. Telese, X. Wang, C.K. Glass, M.G. Rosenfeld, Tyrosinedephosphorylation of H2AX modulates apoptosis and survival decisions,

Nature 458 (2009) 591e596.[73] A. Xiao, H. Li, D. Shechter, S.H. Ahn, L.A. Fabrizio, H. Erdjument-Bromage, et al.,

WSTF regulates the H2A.X DNA damage response via a novel tyrosine kinaseactivity, Nature 457 (2009) 57e62.

[74] T. Liu, Q. Li, Q. Sun, Y. Zhang, H. Yang, R. Wang, et al., MET inhibitor PHA-665752 suppresses the hepatocyte growth factor-induced cell proliferationand radioresistance in nasopharyngeal carcinoma cells, Biochem. Biophys.Res. Commun. 449 (2014) 49e54.

[75] Y. Li, J. Wang, X. Gao, W. Han, Y. Zheng, H. Xu, et al., C-Met targeting enhancesthe effect of irradiation and chemical agents against malignant colon cellsharboring a KRAS mutation, PLoS ONE 9 (2014).

[76] F. De Bacco, P. Luraghi, E. Medico, G. Reato, F. Girolami, T. Perera, et al., In-duction of MET by ionizing radiation and its role in radioresistance andinvasive growth of cancer, J. Natl. Cancer Inst. 103 (2011) 645e661.

[77] J. Michels, I. Vitale, M. Saparbaev, M. Castedo, G. Kroemer, Predictive bio-markers for cancer therapy with PARP inhibitors, Oncogene 33 (2013)3894e3907.

[78] L. Feng, N. Li, Y. Li, J. Wang, M. Gao, W. Wang, et al., Cell cycle-dependentinhibition of 53BP1 signaling by BRCA1, Cell Discov. 1 (2015) 15019.

[79] C.J. Lord, A.N.J. Tutt, A. Ashworth, Synthetic lethality and cancer therapy:lessons learned from the development of PARP inhibitors, Annu. Rev. Med. 66(2015) 455e470.

[80] Z. Feng, J. Zhang, A dual role of BRCA1 in two distinct homologous recombi-nation mediated repair in response to replication arrest, Nucleic Acids Res. 40(2012) 726e738.

[81] Y. Du, H. Yamaguchi, Y. Wei, J.L. Hsu, H.-L. Wang, Y.-H. Hsu, et al., Blocking c-Met-mediated PARP1 phosphorylation enhances anti-tumor effects of PARPinhibitors, Nat. Med. 22 (2016) 194e201.

[82] A. Aly, S. Ganesan, Review BRCA 1, PARP, and 53 BP 1: conditional syntheticlethality and synthetic viability, J. Mol. Cell Biol. (2011) 66e74.

[83] F. Huang, N. Goyal, K. Sullivan, K. Hanamshet, M. Patel, O.M. Mazina, et al.,Targeting BRCA1- and BRCA2-deficient cells with RAD52 small molecule in-hibitors, Nucleic Acids Res. 44 (2016) gkw087.

[84] R.L. Yang, R. Mick, K. Lee, H.L. Graves, K.L. Nathanson, S.M. Domchek, et al.,DCIS in BRCA1 and BRCA2 mutation carriers: prevalence, phenotype, andexpression of oncodrivers C-MET and HER3, J. Transl. Med. 13 (2015) 335.

[85] S.R. Hengel, E. Malacaria, L.F. Da Silva Constantino, F.E. Bain, A. Diaz, B.G. Koch,et al., Small-molecule inhibitors identify the RAD52-ssDNA interaction ascritical for recovery from replication stress and for survival of BRCA2 deficientcells, eLife 5 (2016) e14740.

[86] P. Francica, L. Nisa, D.M. Aebersold, R. Langer, F. Bladt, A. Blaukat, et al.,Depletion of FOXM1 via MET targeting underlies establishment of a DNAdamage-induced senescence program in gastric cancer, Clin. Cancer Res. 22(2016) 5322e5336.