Identification of FES As a Novel Radiosensitizing Target in Human Cancers

Published OnlineFirst October 1, 2019; DOI: 10.1158/1078-0432.CCR-19-0610

CLINICAL CANCER RESEARCH | TRANSLATIONAL CANCER MECHANISMS AND THERAPY

Identiﬁcation of FES as a Novel Radiosensitizing Target in Human Cancers Byoung Hyuck Kim1,2, Yong Joon Kim3,4, Myung-Ho Kim4, Yi Rang Na5, Daun Jung5, Seung Hyeok Seok5, Joon Kim4, and Hak Jae Kim1,6

ABSTRACT ◥ Purpose: The identification of novel targets for developing cell lines. In contrast, FES depletion alone did not significantly synergistic drug–radiation combinations would pave the way to affect cell proliferation without irradiation. An inducible RNAi overcome tumor radioresistance. We conducted cell-based screen- mouse xenograft model verified in vivo radiosensitizing effects. ing of a human kinome siRNA library to identify a radiation-specific FES-depleted cells showed increased apoptosis, DNA damage, – kinase that has a synergistic toxic effect with radiation upon G2 M phase arrest, and mitotic catastrophe after irradiation. FES inhibition and is not essential for cell survival in the absence of depletion promoted radiation-induced reactive oxygen species for- radiation. mation, which resulted in phosphorylation of S6K and MDM2. Experimental Design: Unbiased RNAi screening was performed The radiosensitizing effect of FES knockdown was partially reversed by transfecting A549 cells with a human kinome siRNA library by inhibition of S6K activity. Consistent with the increase in followed by irradiation. Radiosensitizing effects of a target gene and phosphorylated MDM2, an increase in nuclear p53 levels was involved mechanisms were examined. observed, which appears to contribute increased radiosensitivity of Results: We identified the nonreceptor protein tyrosine kinase FES-depleted cells. FES (FEline Sarcoma oncogene) as a radiosensitizing target. The Conclusions: We uncovered that inhibition of FES could expression of FES was increased in response to irradiation. Cell be a potential strategy for inducing radiosensitization in viability and clonogenic survival after irradiation were significantly cancer. Our results provide the basis for developing novel decreased by FES knockdown in lung and pancreatic cancer radiosensitizers.

Introduction radiotherapy, but many of them are nonselective cytotoxic agents that are highly toxic to normal tissues (2). Hypoxic cell radiosensitizers or Radiotherapy is a cornerstone of cancer treatment, and its appli- hypoxic cytotoxins are also limited in use because of side effects (3, 4). cation and frequency are increasing with innovations in radiotherapy The ideal radiosensitizer should not affect the growth and survival of technology. However, the response rate is still unsatisfactory, and cells when they are not receiving radiotherapy and should be able to toxicity is substantial in several cancers, including locally advanced maximize cell death caused by radiotherapy. Although advanced non–small cell lung cancer or pancreatic cancer (1). Radioresistance of radiotherapy technology has enabled highly conformal therapy cancer cells has been the main obstacle, and there are unmet clinical that increases tumor radiation dose while reducing the dose to normal needs for rational approaches to drug–radiotherapy combinations. tissues, the biological efficacy of photon irradiation has not been Several chemotherapies are often used to increase the efficacy of improved (5). Thus, it is necessary to develop novel radiosensitizers that can increase the biological effectiveness of radiotherapy. An important task is to identify therapeutic targets for rational 1 Department of Radiation Oncology, Seoul National University College of radiosensitization. Medicine and Hospital, Seoul, Republic of Korea. 2Department of Radiation Protein kinases are one of the most widely studied therapeutic Oncology, Seoul Metropolitan Government Seoul National University Boramae Medical Center, Seoul, Republic of Korea. 3Department of Ophthalmology, targets because deregulation of kinase function occurs in most can- Institute of Vision Research, Severance Hospital, Yonsei University College of cers (6). In addition, the human kinome is a highly druggable class of Medicine, Seoul, Republic of Korea. 4Graduate School of Medical Science and proteins. However, there have been few efforts to investigate whole Engineering, KAIST, Daejeon, Republic of Korea. 5Department of Microbiology kinases (kinome) as a target for radiosensitization. Here, we present the and Immunology, Institute of Endemic Disease, Seoul National University result of cell-based screening of a human kinome siRNA library to Medical College, Seoul, Republic of Korea. 6Cancer Research Institute, Seoul identify a radiation-specific kinase that has a synergistic toxic effect National University, Seoul, Republic of Korea. with radiation upon inhibition and is not essential for cell survival in Note: Supplementary data for this article are available at Clinical Cancer the absence of radiation. We identified FES as a novel radiosensitizing Research Online (http://clincancerres.aacrjournals.org/). target. FES (FEline Sarcoma oncogene) is known to be an oncogene B.H. Kim and Y.J. Kim contributed equally to this article. that encodes a nonreceptor protein tyrosine kinase and regulates Corresponding Authors: Hak Jae Kim, Seoul National University College of cytoskeletal rearrangements, cell-to-cell interaction, inflammation, Medicine and Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of and innate immunity (7). However, the potential role of radiosensi- Korea. Phone: 82-2-2072-2520; Fax: 82-2-765-3317; E-mail: [email protected]; tization of this orphan kinase is totally unknown to date. and Joon Kim, Graduate School of Medical Science and Engineering, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea. Phone: 82-42-350- 4242; E-mail: [email protected] Materials and Methods – Clin Cancer Res 2020;26:265 73 Cell culture doi: 10.1158/1078-0432.CCR-19-0610 Human pancreatic cancer cell lines MiaPaCa2 and Panc-1, lung 2019 American Association for Cancer Research. cancer cell line A549, and human embryonic kidney 293 cells

AACRJournals.org | 265

Kim et al.

Translational Relevance Results Kinome siRNA library screening identified FES as a novel We identified a novel radiosensitizing target FES using an radiosensitizing target unbiased phenotype-based screening strategy. Our findings shed Unbiased RNAi screening was performed by transfecting A549 new light on the potential strategy for inducing radiosensitization cells with a human kinome siRNA library (715 siRNA pools). Two in human cancer. populations of cells were either nonirradiated or irradiated at 4 Gy, and cell viability was measured after 4 days (Supplementary Fig. S1A). Targets of siRNA pools with a Z-score of surviving fraction at 4 Gy (SF4; cell viability at 4 Gy/cell viability at 0 Gy) of (HEK293T) were purchased from the ATCC. Normal human astrocyte less than -2 were considered as potential hits (Fig. 1A; Supplemen- (NHA) was purchased from the Korean Cell Line Bank. The cells tary Table S1). We then proceeded to secondary screening using fi were cultured at 37 Cina5%CO2 humidi ed chamber. Cells were three or four individual siRNAs for each target and two cancer cell maintained in either DMEM (Welgene) or RPMI1640 medium lines. FES was identified as a radiosensitizing target because two (Welgene) supplemented with 10% FBS and 1% penicillin/strepto- independent FES siRNAs induced a significant reduction in SF4 in mycin, depending on the cell line. both A549 and MiaPaCa2 cells (Supplementary Fig. S1B). Accord- ing to the Human Protein Atlas, most human cancer cells showed RNA interference and chemical reagents moderate to strong cytoplasmic positivity with additional nuclear Cells were transfected with FES-specific siRNAs (CGAGGAUC- positivity of anti-FES immunostaining (8). Several signaling path- CUGAAGCAGUA or GAAAGUGGAUGGCCCAGCG) or nonspe- ways are known to be affected by FES, but, to our knowledge, its role cific negative control siRNA (AllStars Negative Control siRNA, Qia- in radiosensitization has never been addressed. We observed an gen) during 48 hours at 20 nmol/L using Lipofectamine RNAiMAX increase in the total level of FES after irradiation in both A549 and transfection reagent (Invitrogen) in reduced serum medium (Opti- MiaPaCa2 cells, having a peak around 6 hours, which suggests that MEM, Gibco), according to the manufacturer's reverse transfection FES may be functionally related to cellular response to radiation protocol. TAE684 was purchased from Selleckchem. PF-4708671 was (Fig. 1B; Supplementary Fig. S1C). purchased from Sigma-Aldrich. Knockdown of FES increases cancer cell radiosensitivity Kinome-wide siRNA library screening To characterize the effect of FES knockdown on radiosensitivity of Akinome-widesiRNAlibraryfor715humankinaseswas cancer cells, A549, MiaPaCa2, and Panc1 cells were transfected purchased from Dharmacon; the library consists of pools of four with two FES-specific siRNAs or a nonspecific control siRNA. different siRNAs per gene. Polystyrene flat-bottomed 384-well The CCK-8 assay showed that cell viability after irradiation at various plates (Greiner) were spotted with 3 mLof0.25mmol/L siRNA doses was significantly decreased by each FES-specific siRNA in the library pools (62.5 nmol/L for each siRNA) using the BioTek FX three cell lines (Fig. 1C; Supplementary Fig. S2A and S2B). A clono- Laboratory Automation Workstation (Beckman Coulter), and 0.1 genic assay also showed that FES knockdown increased radiation- mL of Lipofectamine RNAiMAX dissolved in 7 mLofOpti-MEM induced death as measured by an increased sensitizer enhancement was mixed in assay plates to perform reverse transfection (total ratio at 10% survival (SER0.1): A549, 1.34 for siFES-1 and 1.30 for siRNA concentration was 15 nmol/L). A549 cells were seeded onto siFES-2; MiaPaCa2, 1.27 for siFES-1 and 1.27 for siFES-2; and Panc1, two pairs of plates (each for 0 and 4 Gy) at 150 cells/well with a final 1.45 for siFES-1 and 1.30 for siFES-2 (Fig. 1D; Supplementary Fig. S2C volume of 50 mL. Irradiation of 4 Gy was performed 48 hours after and S2D). On the other hand, FES knockdown did not radiosensitize siRNA transfection, and cells were further incubated for 96 hours. normal human cells such as NHA and HEK293T, as measured by After that, cell viability was measured after applying Cell Counting clonogenic assay (Supplementary Fig. S2E and S2F). Supplementary Kit-8 (CCK-8) reagent (Dojindo Molecular Technologies) for Fig. S3A indicates an efficient knockdown of FES after transfection of 2 hours. The secondary screening using individual siRNAs of the FES-specific siRNAs. FES knockdown itself did not significantly affect candidate kinases identified in the primary screening was per- cell viability (Supplementary Fig. S3B). To test whether the radio- formed manually in the same manner. sensitizing effect is related to apoptosis promotion, we used an apoptosis assay based on flow cytometry. At 24 hours after irradiation, Study approval FES knockdown moderately increased the apoptotic proportion of All animal procedures were approved by the Institutional Animal Panc1 and MiaPaCa2 cells compared with control group, whereas no Care and Use Committee in Seoul National University Hospital difference was observed without irradiation (Fig. 1E and F; Supple- (SNUH-IACUC No. 18-0001-C1A0), and animals were maintained mentary Fig. S3C). Western blot analysis further confirmed the in the facility accredited by AAALAC International (No. 001169) in augmented cleavage of apoptotic modulators PARP and caspase-3 accordance with Guide for the Care and Use of Laboratory Animals, after irradiation of FES-depleted cells compared with control cells 8th edition, NRC (2010). (Supplementary Fig. S3D). These results indicate that FES knockdown promotes radiation-induced apoptotic cell death. Statistical analysis All values were expressed as the mean SEM, unless otherwise Validation of the radiosensitizing effect of FES loss using FES specified. The mean values between experimental groups were com- knockout A549 cells pared using Student t test. Data analysis and plotting were performed To further confirm the effect of FES inhibition, FES-knockout using GraphPad Prism version 6 (GraphPad Software). (FES-KO) A549 cells were generated using CRISPR/Cas9 system Details about experimental procedures are provided in the Supple- (Fig. 2A; Supplementary Fig. S4). The survival and proliferation mentary Methods. rates between FES-KO and FES-wild type (FES-WT) A549 cells

266 Clin Cancer Res; 26(1) January 1, 2020 CLINICAL CANCER RESEARCH

FES as a Novel Radiosensitizing Target

Figure 1. Kinome siRNA library screening and radiosensitizing effect of FES knockdown in human cancer cells. A, Identification of potential radiosensitizing candidates through siRNA library screening using A549 cells. B, FES protein expression 6 hours after irradiation in MiaPaCa2 and A549 cells. gH2AX was used as a positive control. C, Cell viability of FES or control siRNA transfected A549 cells after each dose of radiation, which were normalized to nonirradiated cells (N ¼ 3). D, Clonogenic cell survival curves of irradiated A549 cells (N ¼ 3). E, Apoptosis assay of FES or control siRNA–transfected cancer cells. F, Quantification of apoptotic cells in E (N ¼ 3). All values are mean SEM. Asterisks indicate statistical significance between control siRNA and FES siRNAs; , P < 0.05; , P < 0.01; and , P < 0.001. were not significantly different when cells were not irradiated Because FES-selective inhibitors are not currently available, we (Fig. 2B). However, radiosensitivity of FES-KO A549 cell was tested the effect of pharmacologic FES inhibition on cell viability increased more than that of FES-WT (Fig. 2C). In addition, the after irradiation using a multikinase inhibitor TAE684, which is clonogenic assay demonstrated a highly effective radiosensitizing known to inhibit FES as well as the well-known target ALK (9). effect of FES inactivation (SER0.1, 2.23; Fig. 2D). Radiation Radiation-induced cell death following a dose of 4 Gy was responses of FES-KO cells in vivo could not be assessed, because significantly increased in the TAE684 group for both A549 (P xenografted FES-KO cells did not form a tumor mass (Supplemen- < 0.001) and MiaPaCa2 (P ¼ 0.012) cells (Fig. 2E). These results tary Fig. S5A and S5B). As a traditional oncogene, FES may also play suggest that FES inhibition can increase the efficacy of a role in tumorigenesis of A549 cells. radiotherapy.

AACRJournals.org Clin Cancer Res; 26(1) January 1, 2020 267

Kim et al.

Figure 2. Effects of FES knockout and inhibition after irradiation. A, Validation of effective FES-KO at protein level. B, Effect of FES-KO on in vitro cell proliferation in A549 cell without irradiation (N ¼ 3). ns, nonsigniﬁcant. C, Cell viability of FES-WT or -KO A549 cells after each dose of radiation normalized to nonirradiated cells in each genotype (N ¼ 3). D, Clonogenic cell survival curves of irradiated A549 FES-KO and -WT cells (N ¼ 3). E, Effect of FES inhibition by TAE684 on the cell viability of irradiated cancer cells (N ¼ 3). Cells were treated with 30 nmol/L TAE684 or 0.1% DMSO for 12 hours, and then irradiated with 4 Gy. CCK-8 assay was performed after 3 days. F, Average xenograft tumor growth curve from three independent experiments (each batch n ¼ 4 per group). Mice were injected with A549 cells containing inducible shCont (nonsilencing shRNA) or shFES at the right hind leg, and treated with or without irradiation (4 Gy three times). Cont, control; IR, irradiation. G, Representative images showing xenograft tumors from shCont and shFES A549 cells. All values are mean SEM. Asterisks indicate statistical signiﬁcance: , P < 0.05; , P < 0.01; and , P < 0.001.

Validation of the radiosensitizing effect of FES depletion using suppressed in A549 cells after xenograft formation by feeding mice inducible FES knockdown system with doxycycline 4 days before irradiation. The FES knockdown group We tested inducible FES knockdown system using a pTRIPZ showed a significant delay in tumor growth after irradiation compared lentiviral vector with tetracycline-inducible shRNA to further validate with the FES-intact control group (Fig. 2F and G). Importantly, the radiosensitizing effect of FES depletion in vivo (Supplementary inducible FES knockdown itself did not affect tumor growth without Fig. S5C and S5D). First, in vitro test showed that doxycycline-induced irradiation. These results verified FES-mediated radiosensitization suppression of FES in A549 cells caused a reduction in cell viability in vivo. after irradiation compared with negative control shRNA (shCont) We analyzed an association between FES gene expression and expression, whereas no significant difference was noted between patient survival using The Cancer Genome Atlas database (Sup- shCont and shFES without doxycycline (Supplementary Fig. S5E). To plementary Fig. S6). There are significant differences in survival of evaluate the therapeutic effect in vivo, FES expression was transiently patients with glioblastoma (HR, 1.5; P ¼ 0.032), low-grade glioma

268 Clin Cancer Res; 26(1) January 1, 2020 CLINICAL CANCER RESEARCH

FES as a Novel Radiosensitizing Target

(HR, 2.3; P < 0.001), and lung squamous cell carcinoma (HR, 1.5; P that FES expression may be related to survival in human cancers, ¼ 0.002) by high FES expression (Supplementary Fig. S6A–S6C). In especially in cancers treated primarily with radiotherapy (such as pancreatic adenocarcinoma and lung adenocarcinoma, there are no brain tumors). differences between the groups. Butitshouldbenotedthatthe number of patients was small and most of them were treated with FES depletion promotes irradiation-mediated cell-cycle arrest surgery or chemotherapy, but not with radiotherapy (Supplemen- and mitotic catastrophe by increasing DNA damage tary Fig. S6D and S6E). When all patients are analyzed together, Our cell-cycle analysis showed that the combination of FES knock- theirsurvivalisclearlydistinguished by the FES expression level down and irradiation increased the sub-G1 fraction and G2–M phase (HR, 1.7; P < 0.001; Supplementary Fig. S6F). These results suggest arrest in A549 and MiaPaCa2 cells (Fig. 3A; Supplementary Fig. S7A).

Figure 3. Effects of FES depletion on irradiation-mediated cell-cycle arrest, mitotic catastrophe, and DNA damage. Cells were transfected with siFES or siCont for 48 hours at 20 nmol/L in all panels. A, Quantification of percentage of cells in each phase of the cell cycle 48 hours after irradiation (N ¼ 3). B, Quantification of the number of mitotic catastrophic cells after various times of irradiation (N ¼ 3). C, Quantification of gH2AX-positive cells after various times of irradiation (N ¼ 3). D, Quantification of comet tail moment after various times of irradiation (N ¼ 3). All experiments used 4-Gy irradiation. All values are mean SEM. Asterisks indicate statistical significance between control siRNA and FES-siRNAs; , P < 0.05; , P < 0.01; and , P < 0.001.

AACRJournals.org Clin Cancer Res; 26(1) January 1, 2020 269

Kim et al.

We next examined whether this arrest results in mitotic catastrophe, statistically significant (Fig. 4F; Supplementary Fig. S8E). This indi- which is a major cell-death mechanism after irradiation. We found that cates that the radiosensitizing effect of FES knockdown could be FES knockdown significantly increased the proportion of cells exhibit- partially reversed by inhibiting S6K activity. Collectively, these results ing mitotic catastrophe after irradiation (Fig. 3B; Supplementary suggest that FES inhibition–mediated radiosensitization involves Fig. S7B). To investigate the effect of FES knockdown on radiation- nuclear p53 accumulation through inhibition of MDM2 activity by induced DNA damage, we analyzed phosphorylated histone H2AX phospho-S6K (Supplementary Fig. S9). (gH2AX) levels. The number of gH2AX-positive cells was significantly increased in the FES knockdown group compared with the control siRNA group at 6 hours following 4-Gy irradiation (Fig. 3C; Supple- Discussion mentary Fig. S7C). Without irradiation, cells demonstrating mitotic Although FES is universally expressed in cancer cells and is known catastrophe and gH2AX positivity were quite few regardless of FES as a signaling mediator downstream of growth factor and cytokine silencing, and no significant difference was noted (Supplementary signaling (15), detailed roles of FES in human cancers remain unclear. Fig. S7D and S7E). A comet assay verified that radiation-induced DNA FES expression was shown to be a poor prognosticator for patients with damage was significantly increased by FES knockdown in both cell prostate or bladder cancer (16, 17). Previous studies have also sug- lines 6 hours after irradiation (Fig. 3D; Supplementary Fig. S7F). gested that FES inhibition may provide therapeutic benefits in kidney Taken together, these results suggest that FES plays a role in the and breast cancers (18, 19). In renal cell carcinoma, downregulation of modulation of the radiation-induced DNA damage response of human FES inhibited cell-cycle progression through the Akt1/NF-kB path- cancer cells. way (18). Qing and colleagues suggested that the JAK3-FES-PLD2 pathway is responsible for the highly proliferative phenotype of breast FES depletion promotes reactive oxygen species production cancer cells (20). However, potential roles of FES in cellular response to and the S6K–MDM2–p53 pathway activation radiation have not previously been studied. Here, we found that the Western blotting demonstrated higher levels of gH2AX in FES- expression of FES in cancer cells is increased in response to irradiation, depleted cells without consistent changes in major DNA-repair suggesting that FES may play a role in the protective response to pathway executors, such as RAD51 and Ku80 (Fig. 4A). Radiation- radiation damage. Using cell culture and mouse xenograft models, we induced foci formation of other DNA repair pathway components further demonstrated the radiosensitizing effect of FES depletion. (pDNA-PK for nonhomologous end joining and MRE11 for homol- Previously, heavy-ion radiation was shown to cause the activation of ogous recombination) was also not suppressed in FES-KO cells 1, 6, Fes and other oncogenes, such as Egfr, NF-kB, and Stat3 in the lung, and 24 hours after irradiation. Rather nonsignificant elevation gut, and small intestinal tissues of healthy mice (21). Targeting EGFR, was found in case of pDNA-PK maybe due to the increased amount NF-kB, and STAT3 is being widely studied for the development of of total DNA damage (Fig. 4B; Supplementary Fig. S8A). Western radiosensitizers (22–25). To our knowledge, this is the first study to blotting of these DNA repair components also confirmed the similar suggest that FES could be a therapeutic target for enhancing results (Supplementary Fig. S8B). Interestingly, the amount of radiosensitivity. reactive oxygen species (ROS) produced after irradiation was We showed that the reduction of ROS scavenging is a potential significantly increased in FES-KO A549 cells (Fig. 4C). The dif- mechanism of radiosensitization by FES inhibition. Because cellular ference was larger when the radiation dose was increased (Fig. 4C). DNA is the major target of radiation and can be damaged not only by FES knockdown also showed similareffectsinA549andMiaPaCa2 radiation but also indirectly by ROS, modulation of the ROS- cells (Supplementary Fig. S8C). Therefore, it is likely that depletion scavenging ability of cancer cells has been considered to be an of FES may contribute to DNA damage by increasing radiation- important strategy to increase radiosensitivity. There has been no induced ROS formation. report of the ability of FES to regulate ROS, but there was one report on ROS-induced S6K phosphorylation is a mediator of radiation- FES-related (FER) kinase, which is a unique cellular homolog of FES induced signal propagation (10, 11). Therefore, we next examined and shares the same domain structure with FES (15). Makovski and whether changes in S6K and its substrate MDM2 are associated with colleagues showed that knockdown of FER increased the level of ROS radiosensitization after FES depletion. Previously, it was reported that in colon carcinoma cells (26). FER has been also studied as a new FES increases S6K activation and knockdown of FES decreases phos- therapeutic target for some cancers (27, 28). These results support the phorylation of S6K (12). Unexpectedly, the combination of FES reason for further study of the role of the c-FES family of protein knockdown and irradiation caused an increase in phosphorylation of kinases in radiation biology. Because cancer cells are heterogeneous S6K and MDM2 (Fig. 4D). Phosphorylation of MDM2 was also and possess a diverse range of mutations and alterations, the effects of accompanied by p53 accumulation in the nucleus (Fig. 4D), which FES inhibition may differ between individual cell types. Further is known to play a key role in the execution of apoptosis after investigations are warranted to understand the role and mechanism irradiation (13, 14). Similar results were confirmed using FES-KO of FES in the regulation of ROS in various cancers suitable for A549 cells (Supplementary Fig. S8D). Therefore, elevated activity of the radiotherapy. S6K/MDM2/p53 axis may be one of the mechanisms by which FES Kinase activity of FES seems to be regulated by coiled-coil– knockdown synergizes with irradiation to cause cell-cycle arrest and dependent oligomerization and subsequent autophosphorylation, apoptosis. but its general substrates and upstream regulators have not been To further support that ROS-mediated activation of S6K contri- revealed (29). Our results suggest that the S6K/MDM2/p53 axis butes to increased radiosensitivity of FES-depleted cells, we examined acting downstream of ROS is a key mediator of the radiosensitizing the effect of S6K inhibition. Indeed, treatment of the selective S6K effect of FES inhibition. However, we did not fully investigate inhibitor PF-4708671 in FES-depleted cells attenuated the induction of cellular signaling pathways that might be associated with the PARP cleavage and gH2AX expression (Fig. 4E), and slightly increase in irradiation-mediated apoptosis, cell-cycle arrest, mitotic increased clonogenic survival and cell viability in irradiated A549 and catastrophe, and DNA damage in FES-depleted cells. An unexam- MiaPaCa2 cells, although differences in some phenotypes were not ined DNA repair pathway could be also affected. Previously, it has

270 Clin Cancer Res; 26(1) January 1, 2020 CLINICAL CANCER RESEARCH

FES as a Novel Radiosensitizing Target

Figure 4. FES depletion promotes ROS production and affects the S6K-MDM2-p53 pathway. A, DNA-repair pathway executors and gH2AX protein levels in control and FES knockdown (KD) cells 6 hours after 4-Gy irradiation. B, Quantification of the pDNA-PK–positive cells after various times of irradiation (N ¼ 3). C, Representative results of flow cytometry analysis of total cellular ROS contents in A549 cells 4 hours after irradiation. D, Western blot analysis of the S6K/MDM2/p53 pathway. Cells with or without FES siRNAs were irradiated with 4 Gy, and then protein was extracted 6 hours after irradiation. In the bottom plot, nuclear fractions (nuc) 24 hours after irradiation were used. E, Western blot analysis of selective S6K inhibition by PF-4708671 with or without FES siRNAs transfection. Cells were irradiated with 4 Gy, and then protein samples were extracted after 6 hours. F, Clonogenic cell survival curves of irradiated cells with or without S6K inhibition (N ¼ 3). All values are mean SEM. been shown that DNA repair can be suppressed via S6K-dependent but it seems like the role of FES in tumorigenesis differs according to MRE11 downregulation, which is a component of the DNA dam- the tumor type as well as diverse circumstances. This might reflect the age–response complex (30). Althoughouranalysesdidnotreveal complexity of FES interaction with multiple cellular pathways which noticeable changes in the DNA repair pathway, an activation of S6K has not been completely revealed. We were not able to evaluate the after irradiation in FES-depleted cells could also partially contribute radiosensitivity of FES-KO xenografts, because FES was apparently to radiosensitizing effects through DNA repair disturbance. required for tumorigenesis of A549 cells in vivo. FER expression is FES has been mentioned in the previous literature as an oncogene as known to be associated with anoikis resistance, and its high expression well as a tumor suppressor (31, 32). There are not enough studies yet, predicts poor prognosis (33). Thus, it is likely that FES-KO A549 cells

AACRJournals.org Clin Cancer Res; 26(1) January 1, 2020 271

Kim et al.

failed to form xenograft tumor because they were not free from anoikis Analysis and interpretation of data (e.g., statistical analysis, biostatistics, in living organisms. In contrast, transient depletion of FES did not computational analysis): B.H. Kim, Y.J. Kim, M.-H. Kim, Y.-R. Na, D. Jung, interfere with xenograft tumor formation, suggesting that the depen- S.-H. Seok, H.J. Kim fi Writing, review, and/or revision of the manuscript: B.H. Kim, Y.J. Kim, Y.-R. Na, dence of cancer cells on FES activity is speci c to the stage of tumor J. Kim, H.J. Kim development. Administrative, technical, or material support (i.e., reporting or organizing data, In conclusion, we identified a novel radiosensitizing target FES constructing databases): Y.J. Kim, H.J. Kim using an unbiased phenotype-based screening strategy. If a specific Study supervision: J. Kim, H.J. Kim inhibitor of FES is developed, it is expected to increase the effectiveness of radiotherapy and be a great help in expanding the indications for Acknowledgments radiotherapy. Clinically usable radiosensitizers could have great eco- We gratefully acknowledge Soo Yeon Seo and Da Young Park for expert technical nomic value and reduce socioeconomic costs. Our study might also support. This research was supported by the Basic Science Research Program through help oncologists to design further research to explore the radioresis- the National Research Foundation of Korea funded by the Ministry of Science and tance of malignant tumors. ICT (2017R1A2B3005208, to J. Kim). This work was also supported by grants from the Ministry of Science and ICT (grant numbers NRF2017M2A2A7A01070925 and Disclosure of Potential Conflicts of Interest 2016R1D1A1B03934691, to H.J. Kim). No potential conflicts of interest were disclosed. The costs of publication of this article were defrayed in part by the payment of page advertisement ’ charges. This article must therefore be hereby marked in accordance Authors Contributions with 18 U.S.C. Section 1734 solely to indicate this fact. Conception and design: B.H. Kim, Y.J. Kim, J. Kim, H.J. Kim Development of methodology: B.H. Kim, Y.J. Kim Acquisition of data (provided animals, acquired and managed patients, provided Received February 19, 2019; revised August 29, 2019; accepted September 27, 2019; facilities, etc.): B.H. Kim, Y.J. Kim, Y.-R. Na, S.-H. Seok, H.J. Kim published first October 1, 2019.

References 1. Bradley JD, Paulus R, Komaki R, Masters G, Blumenschein G, Schild S, et al. 15. Smithgall TE, Rogers JA, Peters KL, Li J, Briggs SD, Lionberger JM, et al. The c- Standard-dose versus high-dose conformal radiotherapy with concurrent and Fes family of protein-tyrosine kinases. Crit Rev Oncog 1998;9:43–62. consolidation carboplatin plus paclitaxel with or without cetuximab for patients 16. Miyata Y, Watanabe S, Matsuo T, Hayashi T, Sakai H, Xuan JW, et al. with stage IIIA or IIIB non-small-cell lung cancer (RTOG 0617): a randomised, Pathological significance and predictive value for biochemical recurrence of two-by-two factorial phase 3 study. Lancet Oncol 2015;16:187–99. c-Fes expression in prostate cancer. The Prostate 2012;72:201–8. 2. Linam J, Yang L-X. Recent developments in radiosensitization. Anticancer Res 17. Asai A, Miyata Y, Takehara K, Kanda S, Watanabe S-I, Greer PA, et al. 2015;35:2479–85. Pathological significance and prognostic significance of FES expression in 3. Fazekas J, Pajak TF, Wasserman T, Marcial V, Davis L, Kramer S, et al. Failure of bladder cancer vary according to tumor grade. J Cancer Res Clin Oncol misonidazole-sensitized radiotherapy to impact upon outcome among stage III- 2018;144:21–31. IV squamous cancers of the head and neck. Int J Radiat Oncol Biol Phys 1987;13: 18. Kanda S, Miyata Y, Kanetake H, Smithgall TE. Downregulation of the c-Fes 1155–60. protein-tyrosine kinase inhibits the proliferation of human renal carcinoma cells. 4. Lee DJ, Cosmatos D, Marcial VA, Fu KK, Rotman M, Cooper JS, et al. Results of Int J Oncol 2009;34:89–96. an RTOG phase III trial (RTOG 85-27) comparing radiotherapy plus etanidazole 19. Zhang S, Chitu V, Stanley ER, Elliott BE, Greer PA. Fes tyrosine kinase expression with radiotherapy alone for locally advanced head and neck carcinomas. Int J in the tumor niche correlates with enhanced tumor growth, angiogenesis, Radiat Oncol Biol Phys 1995;32:567–76. circulating tumor cells, metastasis, and infiltrating macrophages. Cancer Res 5. Kim JH, Jenrow KA, Brown SL. Novel biological strategies to enhance the 2011;71:1465–73. radiation therapeutic ratio. Radiat Oncol J 2018;36:172–81. 20. Ye Q, Kantonen S, Henkels KM, Gomez-Cambronero J. A new signaling pathway 6. Futreal PA, Coin L, Marshall M, Down T, Hubbard T, Wooster R, et al. A census (JAK-Fes-phospholipase D) that is enhanced in highly proliferative breast cancer of human cancer genes. Nat Rev Cancer 2004;4:177–83. cells. J Biol Chem 2013;288:9881–91. 7. Greer P. Closing in on the biological functions of Fps/Fes and Fer. Nat Rev Mol 21. Aravindan N, Aravindan S, Manickam K, Natarajan M. High energy particle Cell Biol 2002;3:278–89. radiation-associated oncogenic transformation in normal mice: insight into the 8. Uhlen M, Zhang C, Lee S, Sjostedt€ E, Fagerberg L, Bidkhori G, et al. A pathology connection between activation of oncotargets and oncogene addiction. Sci Rep atlas of the human cancer transcriptome. Science 2017;357:6352. 2016;6:37623. 9. Hellwig S, Miduturu CV, Kanda S, Zhang J, Filippakopoulos P, Salah E, et al. 22. Choi EJ, Ryu YK, Kim SY, Wu HG, Kim JS, Kim IH, et al. Targeting epidermal Small-molecule inhibitors of the c-Fes protein-tyrosine kinase. Chem Biol 2012; growth factor receptor-associated signaling pathways in non-small cell lung 19:529–40. cancer cells: implication in radiation response. Mol Cancer Res 2010;8:1027–36. 10. Contessa JN, Hampton J, Lammering G, Mikkelsen RB, Dent P, Valerie K, et al. 23. Xu Y, Fang F, St Clair DK, Sompol P, Josson S, St Clair WH. SN52, a novel nuclear Ionizing radiation activates Erb-B receptor dependent Akt and p70 S6 kinase factor-kappaB inhibitor, blocks nuclear import of RelB:p52 dimer and sensitizes signaling in carcinoma cells. Oncogene 2002;21:4032–41. prostate cancer cells to ionizing radiation. Mol Cancer Ther 2008;7:2367–76. 11. Huang C, Li J, Ke Q, Leonard SS, Jiang B-H, Zhong X-S, et al. Ultraviolet-induced 24. Huang C-Y, Tai W-T, Wu S-Y, Shih C-T, Chen M-H, Tsai M-H, et al. Dovitinib phosphorylation of p70(S6K) at Thr(389) and Thr(421)/Ser(424) involves acts as a novel radiosensitizer in hepatocellular carcinoma by targeting SHP-1/ hydrogen peroxide and mammalian target of rapamycin but not Akt and atypical STAT3 signaling. Int J Radiat Oncol Biol Phys 2016;95:761–71. protein kinase C. Cancer Res 2002;62:5689–97. 25. Han TJ, Cho BJ, Choi EJ, Kim DH, Song SH, Paek SH, et al. Inhibition of STAT3 12. Voisset E, Lopez S, Dubreuil P, De Sepulveda P. The tyrosine kinase FES is an enhances the radiosensitizing effect of temozolomide in glioblastoma cells essential effector of KITD816V proliferation signal. Blood 2007;110:2593–9. in vitro and in vivo. J Neurooncol 2016;130:89–98. 13. Liu B, Bhatt D, Oltvai ZN, Greenberger JS, Bahar I. Significance of p53 dynamics 26. Makovski A, Yaffe E, Shpungin S, Nir U. Down-regulation of Fer induces ROS in regulating apoptosis in response to ionizing radiation, and polypharmaco- levels accompanied by ATM and p53 activation in colon carcinoma cells. logical strategies. Sci Rep 2014;4:6245. Cell Signal 2012;24:1369–74. 14. Lai KP, Leong WF, Chau JFL, Jia D, Zeng L, Liu H, et al. S6K1 is a multifaceted 27. Oneyama C, Yoshikawa Y, Ninomiya Y, Iino T, Tsukita S, Okada M. Fer tyrosine regulator of Mdm2 that connects nutrient status and DNA damage response. kinase oligomer mediates and amplifies Src-induced tumor progression. Onco- EMBO J 2010;29:2994–3006. gene 2016;35:501–12.

272 Clin Cancer Res; 26(1) January 1, 2020 CLINICAL CANCER RESEARCH

FES as a Novel Radiosensitizing Target

28. Zheng Y, Sethi R, Mangala LS, Taylor C, Goldsmith J, Wang M, et al. Tuning 31. Sangrar W, Zirgnibl RA, Gao Y, Muller WJ, Jia Z, Greer PA. An identity microtubule dynamics to enhance cancer therapy by modulating FER-mediated crisis for fps/fes: oncogene or tumor suppressor? Cancer Res 2005;65: CRMP2 phosphorylation. Nat Commun 2018;9:476. 3518–22. 29. Tagliaﬁco E, Siena M, Zanocco-Marani T, Manfredini R, Tenedini E, Montanari 32. Olvedy M, Tisserand JC, Luciani F, Boeckx B, Wouters J, Lopez S, et al. M, et al. Requirement of the coiled-coil domains of p92(c-Fes) for nuclear Comparative oncogenomics identiﬁes tyrosine kinase FES as a tumor suppressor localization in myeloid cells upon induction of differentiation. Oncogene 2003; in melanoma. J Clin Invest 2017;127:2310–25. 22:1712–23. 33. Ivanova IA, Vermeulen JF, Ercan C, Houthuijzen JM, Saig FA, Vlug EJ, et al. 30. Piscitello D, Varshney D, Lilla S, Vizioli MG, Reid C, Gorbunova V, et al. AKT FER kinase promotes breast cancer metastasis by regulating a6- and b1- overactivation can suppress DNA repair via p70S6 kinase-dependent down- integrin-dependent cell adhesion and anoikis resistance. Oncogene 2013;32: regulation of MRE11. Oncogene 2018;37:427–38. 5582–92.

AACRJournals.org Clin Cancer Res; 26(1) January 1, 2020 273

Identification of FES as a Novel Radiosensitizing Target in Human Cancers

Byoung Hyuck Kim, Yong Joon Kim, Myung-Ho Kim, et al.

Clin Cancer Res 2020;26:265-273. Published OnlineFirst October 1, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-19-0610

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/10/01/1078-0432.CCR-19-0610.DC1

Cited articles This article cites 33 articles, 9 of which you can access for free at: http://clincancerres.aacrjournals.org/content/26/1/265.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/26/1/265. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.