Inhibition of JNK Sensitizes Hypoxic Colon CancerCells to DNA-Damaging AgentsIrina A. Vasilevskaya1, Muthu Selvakumaran1, Lucia Cabal Hierro1, Sara R. Goldstein2,Jeffrey D.Winkler2, and Peter J. O'Dwyer1
Abstract
Purpose: We showed previously that in HT29 colon cancercells, modulation of hypoxia-induced stress signaling affectsoxaliplatin cytotoxicity. To further study the significance ofhypoxia-induced signaling through JNK, we set out to investigatehowmodulation of kinase activities influences cellular responsesof hypoxic colon cancer cells to cytotoxic drugs.
Experimental Design: In a panel of cell lines, we investigatedeffects of pharmacologic and molecular inhibition of JNK onsensitivity to oxaliplatin, SN-38, and 5-FU. Combination studiesfor the drugs and JNK inhibitor CC-401 were carried out in vitroand in vivo.
Results: Hypoxia-induced JNK activation was associated withresistance to oxaliplatin. CC-401 in combination with chemo-therapy demonstrates synergism in colon cancer cell lines,although synergy is not always hypoxia specific. A more detailed
analysis focused on HT29 and SW620 (responsive), and HCT116(nonresponsive) lines. In HT29 and SW620 cells, CC-401 treat-ment results in greater DNA damage in the sensitive cells. In vivo,potentiation of bevacizumab, oxaliplatin, and the combinationby JNK inhibition was confirmed in HT29-derived mouse xeno-grafts, in which tumor growth delay was greater in the presence ofCC-401. Finally, stable introduction of a dominant negativeJNK1, but not JNK2, construct into HT29 cells rendered themmore sensitive to oxaliplatin under hypoxia, suggesting differinginput of JNK isoforms in cellular responses to chemotherapy.
Conclusions: These findings demonstrate that signalingthrough JNK is a determinant of response to therapy in coloncancer models, and support the testing of JNK inhibition tosensitize colon tumors in the clinic. Clin Cancer Res; 21(18); 4143–52.�2015 AACR.
IntroductionA salient characteristic of solid tumors is the presence of
hypoxic regions, which form mostly as a result of accelerated cellgrowth and increased distance to blood vessels. Through neovas-cularization, oxygen influx to hypoxic regions can be restored,creating a complex tumor microenvironment, where continuoushypoxia can be followed by reoxygenation (1, 2). This phenom-enon drives several processes associated with poor prognosis,including a more aggressive phenotype, resistance to chemother-apy and radiation, and increased genetic instability (3). A majortranscriptional consequence of hypoxia is activation of hypoxia-inducible factors (HIF), which regulate multiple pathwaysinvolved in metabolic reprogramming, cell death, angiogenesis,epithelial to mesenchymal transition (EMT), andmaintenance ofcancer stem cells (4, 5). Thus, hypoxia is a legitimate target toaugment outcomes of cancer therapy (6, 7). HIF, however, is notthe sole factormediating tumor responses tohypoxia. Thebreadthof these responses reflects involvement of various signaling path-
ways, resulting in activation of other major transcription factors,such as NF-kB, CREB, p53, and AP-1 (8).
The AP-1 transcription factor consists of members of Jun, Fos,ATF, and MAF protein families. By forming various homo- andheterodimers, AP-1 activates transcription of a broad range ofgenes involved in cell proliferation, differentiation, apoptosis,and other cellular functions (9). Activation of AP-1 occurs mostlythrough transcriptional induction and/or activating phosphory-lation byMAPKs, which are the proximalmembers of three-tieredsignal transduction cascades mediating cellular responses to var-ious external and internal stimuli (10). Upstream activators ofMAPKs, MAPK kinases (MAPKKs or MKKs), belong to a classof tyrosine protein kinases, and activate their correspondingtargets after being induced, in turn, by multiple MAPKK kinases(MAPKKKs, MEKKs, or MKKKs; ref. 11). Mitogenic signals arepreferentially relayed though MEK1 and MEK2, which activateERK1/2 (12), whereas stress signaling is mediated by MKK3/MKK6 which phosphorylates p38 MAPK, and MKK4/MKK7which activates JNK1/2/3 (13, 14). Signal transduction throughMAP kinases is important for normal cell function, but in cancerERKs, p38MAPKs and JNKs can demonstrate both oncogenic andcancer suppressive features (15–17).
In our earlier work, we found that hypoxic induction of AP-1 incolon cancer cell lines was associated with resistance to chemo-therapy (18–20). We demonstrated that in HT29 colon adeno-carcinoma, hypoxia activates p38 MAPK and JNK, whereas ERK1and ERK2 are constitutively active, and cJun/AP-1 activation ismediated predominantly by JNK (20). We then studied upstreamactivators of JNK, MKK4, and MKK7, and showed not only theactivation of both under hypoxia, but that selective downregu-lation of this signaling exerts differing effects on oxaliplatin
1Abramson Cancer Center, University of Pennsylvania, Philadelphia,Pennsylvania. 2Department of Chemistry, University of Pennsylvania,Philadelphia, Pennsylvania.
Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).
Corresponding Author: Irina A. Vasilevskaya, University of Pennsylvania, 1020BRBII/III, 421 Curie Blvd, Philadelphia PA 19104. Phone: 215-573-7300; Fax: 215-573-7049; E-mail: [email protected]
cytotoxicity: MKK4 deficiency results in higher sensitivity tooxaliplatin, whereas downregulation of MKK7 renders HT29 cellsmore resistant to the drug in vitro and in vivo (21).These datasuggested that hypoxic signaling through MKK4 could contributeto oxaliplatin resistance, while uninterrupted signaling to AP-1through the MKK7/JNK module is essential for oxaliplatin cyto-toxicity in hypoxic HT29 cells. Accordingly, targeting separatecomponents of JNK signaling pathway or its downstream targetscould clarify conflicting data concerning the role of JNK inchemotherapeutic resistance in general (22) and yield novelapproaches to enhance oxaliplatin cytotoxicity in particular.
Here, we study the effects of JNK inhibition on sensitivity tooxaliplatin, SN-38, and 5-FU in a panel of 12 colon cancer celllines. Our data show that inhibition of JNK by CC-401 enhancescytotoxicity of the chemotherapy in vitro. Sensitization to oxali-platinwas confirmed in vivo, anddownregulation of JNK1, but notJNK2, by dominant negative constructs rendered hypoxic HT29cells more sensitive to oxaliplatin. Our findings support furthertesting of JNK inhibitors in the clinic.
Materials and MethodsCells and reagents
All human colon cancer cell lines were from ATCC.HCT116p53�/� was kindly provided by Dr. B. Vogelstein (JohnHopkins Kimmel Cancer Center, Baltimore, MD) and BE cellswere from Dr. B. Giovanella (St. Joseph's Hospital CancerResearch Laboratory, Houston, TX.) Cells were grown in DMEMmedium supplemented with 10% FBS and antibiotic-antimycoticreagent (Invitrogen). SP600125 was purchased from Biomol,oxaliplatin from LKT Labs, and 7-Ethyl-10-hydroxycamptothecin(SN-38) and 5-fluorouracil (5-FU) from Sigma-Aldrich. CC-401was synthesized in the laboratory of Dr. Jeffrey D. Winkler,Department of Chemistry of the University of Pennsylvania(Philadelphia, PA).
Plasmids and isolation of stably transfected cell linesThe HA-tagged dominant negative mutants of JNK1 and JNK2
(HA-JNK1-APF and HA-JNK2-APF, respectively) cloned into ret-roviral pLNCX vector were kindly provided by Dr. Tomas Berl
(University of Colorado, Denver, CO). To isolate cell lines stablyexpressing empty vector and dnJNK1 or dnJNK2, HT29 cells weretransfected using Fugene 6 transfection reagent (Roche AppliedSciences), according to the manufacturer's recommendation andcultivated at low density inmedia containing G418 (0.75mg/mL,Invitrogen); surviving colonies were isolated, propagated, andassessed for expression of tag protein.
Hypoxic treatmentCells were exposed to acute hypoxia in an anaerobic chamber
(Forma Scientific, Inc.) filled with a gas mixture consisting of 5%CO2, 9% H2, and 86% N2. Oxygen content (below 0.5%) wasmonitoredbyPROOX110oxygen sensor (BioSpherix). Cellswereplated in 100-mmglass Petri dishes to adensity of 2�106 cells perdish and subjected to hypoxia within 36 hours. The cells wereharvested at various time points for further experiments.
Protein extract preparation and Western blottingTotal protein extracts were prepared as described in ref. (21),
using cell lysis buffer (Cell Signaling Technology), supplementedwith complete protease inhibitor cocktail (Roche) and 1 mmol/LPMSF (Sigma). Cells were lysed inside the hypoxia chamber,followed by incubation in a shaker for 30 minutes at 4�C, andcentrifugation. Protein concentration of cleared cellular extractswas measured using the Bio-Rad Protein Assay (Bio-Rad).
For protein electrophoresis in SDS-polyacrylamide gels, pro-tein extracts were used in amounts of 10 mg per lane. Westernblotting was carried out according to standard procedures, usinghorseradish peroxidase-conjugated secondary antibodies (SantaCruz Biotechnology) and the ECLþPlus detection system (Amer-sham). Results were analyzed with BioSpectrum 810 ImagingSystem using VisionWorksLS Image Acquisition and AnalysisSoftware (UVP).Theprimary antibodies against actin,Chk1,Chk2were from Santa Cruz Biotechnology; the remaining antibodieswere purchased fromCell Signaling Technology. For IHC, primaryantibodies against CD31 (Abcam) and against Ki-67 (Dako) werealso used.
Cytotoxicity assays and calculation of combination indicesFor assessment of cytotoxicity, cells were plated in 96-well
plates (2,000 cells per well), and 24 hours later, various concen-trations of chemotherapeutic drugs alone or in combination wereadded, immediately before hypoxic exposure for 24 hours, fol-lowed by cultivation in oxic conditions for an additional 48hours. Cytotoxicitywasmeasuredusing a standardMTTassay. TheIC50 values shown are the means of at least three independentexperiments done in triplicate. The combination index analysiswas performed using the median effect method by Chou andTalalay (23, 24).
Analysis of DNA damage inductionTo study the induction of DNA damage, the levels of phos-
phorylated H2AX were measured. Cells were seeded in 6-wellplates (2 � 105 cells/plate) and 24 hours later subjected tohypoxia and/or drug treatment for 24 hours. Cells were collected(including any floating cells in the culture medium), washedtwice with PBS, permeabilized on ice for 15 minutes with 1%formaldehyde, resuspended in 70% ethanol, and kept at �20�Covernight. Next, after two washes with PBS and one with 1% BSA-PBS, cells were resuspended in 100 mL of 1% BSA-PBS together
Translational Relevance
The combination of cytotoxic chemotherapy and bevacizu-mab is the backbone of therapy for advanced colorectal cancer.Understanding the basis of resistance is the key to develop-ment of newer and more effective therapies. Our previouswork suggests that the induction of hypoxia by bevacizumabcontributes to its interaction with chemotherapy, and mayincrease cell kill while promoting the acquisition ofresistance. We found that inhibition of the JNK pathway atthe SEK kinase level increased the efficacy of both bevacizu-mab and oxaliplatin. In this paper, we show that directinhibition of JNK sensitizes colorectal cancer cells of varyinggenetic background to hypoxia and cytotoxic drugs, and thatJNK inhibition may be a promising therapeutic approach inthis disease. Furthermore, molecular disruption of JNK iso-forms suggests that additional specificity might result fromselective inhibition of JNK1.
Vasilevskaya et al.
Clin Cancer Res; 21(18) September 15, 2015 Clinical Cancer Research4144
with 5 mL of the p-H2AX (ser139) antibody (Alexa Fluor 647labeled, BD Pharmingen) and incubated at RT for 2 hours in thedark before the addition of 300 mL of 1%BSA-PBS. The percentageof positive cells was then determined by flow cytometry (BDFACSCalibur), and results analyzed using FlowJo software(TreeStar).
Tumor growth and antiangiogenic therapyTo assess the efficacy of JNK signaling inhibition by CC-401
in antiangiogenic and oxaliplatin combination therapy in amouse xenograft model, adult (8–10 weeks of age) femaleSCID mice (C.B.17 SCID) were used. To generate tumors, HT29
cells (1 � 106 cells) were injected subcutaneously into the leftflank of the mice. When the tumors reached approximately 200mm3, mice were divided into eight groups (eight mice pergroup) for treatment with bevacizumab (Genentech, Inc.),oxaliplatin, CC401, and the appropriate combinations of bev-acizumab, oxaliplatin, and CC-401. Mice in the bevacizumabtreatment group received 5 mg/kg of bevacizumab by intraper-itoneal injection every 3 days for 21 days. The oxaliplatintreatment group was injected intraperitoneally with 5 mg/kgoxaliplatin per week for 2 weeks. The CC-401 treatment groupwas injected intraperitoneally 25 mg/kg for every 3 days. Thecombination treatment groups received bevacizumab (every 3days, 5 mg/kg), oxaliplatin (weekly for 2 weeks, 5 mg/kg), andCC-401 (every 3 days, 25 mg/kg). The control group receivedsaline intraperitoneally. Tumor volume and body weight weremeasured every 3 days. Tumor volume was calculated using theformula V ¼ AB2/2, where A is the largest diameter and B isthe smallest diameter. Tumor growth delay was calculated asthe difference in the time for control and treated tumors togrow from 200 to 800 mm3. For tumor growth delay calcula-tions, mice were continued to receive treatments until thetumor volume reached 800 mm3. For IHC, mice were sacrificedafter treatments on day 9 for tumor processing and staining. Allanimal experiments were performed according to an approvedUniversity of Pennsylvania IACUC (Institutional Animal Careand Use Committee).
Tissue IHCIHC was conducted using the antigen retrieval protocol
followed by primary antibody incubation previously described
Figure 1.Hypoxia induces signaling through JNKand chemoresistance in colon cancer cell lines. A,Western blot analyses of c-Jun phosphorylation in hypoxic colon cancer celllines. Lower band shows actin content, as a control of equal loading. B, responses in a panel of colon cancer cell lines to oxaliplatin, SN-38, and 5-FU under hypoxia:shown are the ratios of IC50 under hypoxia to that of the oxic control; JNK activation presented as a peak value of p-c-Jun during hypoxia as compared withoxic control. Highlights mark cell lines with more than twofold increase in IC50 of the corresponding drug or JNK activation. � , P < 0.05.
Table 1. Combination of JNK inhibitor CC-401 with chemotherapy is synergisticin colon cancer cell lines
Oxaliplatin SN-38 5-FUCell line CI oxic CI hypa CI oxic CI hypb CI oxic CI hypb
NOTE: Cytotoxic interactions of JNK inhibitor CC-401 with oxaliplatin, SN-38,and5-FU in colon cancer cellswere evaluated inMTT-basedassays. Combinationindices (CI50) were calculated based on the Chou–Talalay method usingCompuSyn software. Presented are the average values from three independentexperiments in triplicate. Similar experiments were carried out with SP600125(Supplementary Table S3).aP < 0.001, significant when compared with control (oxic) group.bP values are not significant when compared with the control group.
JNK Inhibition Potentiates Therapy in Colon Cancer Models
www.aacrjournals.org Clin Cancer Res; 21(18) September 15, 2015 4145
(25, 26). Rabbit anti-human p-JNK and p-cJun antibodies wereused to detect the expression of p-JNK and p-cJun in tumors. Forblood vessels, Ki-67 markers, and p-H2AX detection, antibodiesagainst CD31, Ki-67, and p-H2AX were used. IHC images werevisualized using a Leica DMRBE upright microscope with QIma-gingMicroPublisher 5.0 RTV color camera with objective lens PLFluoTAR 20x/0.5 (Leica). Images were captured using iVisionacquisition software, which were processed with Adobe Photo-shop software (Adobe Systems). Tumor blood vessels densitieswere determined by counting per �40 high power field asdescribed earlier (25, 26). For quantitative analysis, ImageJ soft-ware, measuring the intensity of staining through thresholdanalysis, coupled with the Color Deconvolution plug-in was used
to quantify immunoreactivity in xenograft samples (http://www.mecourse.com/landinig/software/cdeconv/cdeconv.html).
Statistical analysisData were analyzed using the Student t test and ANOVA
where appropriate. A P value less than 0.05 was accepted as astatistically significant difference when compared with corre-sponding control, and marked with an asterisk in figures. Toassess effects of the drug's combinations between oxic andhypoxic conditions, combinatorial indices (CI) were evaluatedusing two-way ANOVA models. In graphs, values are presentedas mean plus SD.
Figure 2.Hypoxia causesDNAdamage in a cell-specificmanner. A, cellswere subjected to hypoxiawith orwithout CC-401 at the doseof 1� IC50 (for HT29, HCT116, andSW620,respectively) for 6 or 24 hours, followed by Western blot analysis to assess c-Jun phosphorylation, or activation of Chk1 and Chk2 as indicators of DNA damagesignaling. B, graph presents effects of CC-401 on c-Jun phosphorylation and activation of DNA-damage response under hypoxia, based on calculation ofband densities normalized to actin; average values from at least two independent experiments are shown. C, cell lines were subjected to 24 hours of hypoxia with orwithout oxaliplatin at the dose of IC50 for oxic condition in the absence (black bars) or presence (striped bars) of CC-401. Percentages of p-H2AX–positive cells wereestablished in three independent experiments and plotted as fold-increase compared with oxic control. � , P < 0.05.
Vasilevskaya et al.
Clin Cancer Res; 21(18) September 15, 2015 Clinical Cancer Research4146
ResultsHypoxia induces selective signaling through JNK and increasesresistance to chemotherapy in a panel of colon cancer cell lines
We showed previously that induction of the JNK pathway inhypoxic HT29 colon adenocarcinoma cells was functionally asso-ciatedwith resistance, and that cells could be rendered sensitive bydownregulation of MKK4 (SEK1; refs. 19, 21). To extend observa-tions on this pathway, we evaluated induction of stress signalingthrough JNK by hypoxia in a panel of 12 colon cancer cell lines.Cells were subjected to hypoxia for 1, 3, 5, or 24 hours, andphosphorylation of c-Jun, as a marker of JNK induction, wasmeasured by Western blotting (Fig. 1A). The levels of c-Junphosphorylation, calculated by comparing band densities to theone from control sample, were established in at least two inde-pendent experiments for all of cell lines, and presented in Sup-plementary Table S1. We found that hypoxia induces signalingthrough JNK in a majority of cell lines, to varying degrees, andwith cell-specific temporal characteristics. Induction of signalingin this pathway is an early response, which subsides by 24hours ofhypoxic exposure in a majority of cell lines; therefore, we desig-nated lines with increased content of phospho-cJun by 5 hours ofhypoxia as positive for hypoxia-induced JNK signaling. In thesame cellular panel, we also assessed sensitivity to oxaliplatin, SN-38, and 5-FU under normal or hypoxic conditions using MTTassays. Our data show a wide variety of responses (IC50 values forall drugs shown in Supplementary Table S2): under hypoxiamore
than a2-fold increase in resistance tooxaliplatinwas observed in 6cell lines, sensitivity to SN-38 diminished in 9 cell lines, and 3 celllines becamemore resistant to 5-FU (Fig. 1B). Figure 1B illustratesthe relationship of hypoxia-induced resistance to chemotherapyand induction of JNK signaling by hypoxia, as assessed by c-Junphosphorylation. Evidently, hypoxic resistance to chemotherapyis not associated with activation of JNK signaling in DLD1,HCT15, HCT116, and its p53-null derivative, HCT116p53�/�,but in the other eight cell lines, induction of the JNK pathway byhypoxia is associatedwith resistance. Therefore, wewished to findout whether inhibition of JNK would influence the efficacy ofchemotherapy.
Pharmacologic inhibition of JNK sensitizes colon cancer celllines to chemotherapeutic drugs
To investigate the effects of JNK inhibition on drug resistanceunder hypoxia, we carried out MTT assays using the small-mol-ecule JNK inhibitors CC-401 and SP600125 in combinationswithoxaliplatin, SN-38, and 5-FU, followed by calculation of combi-nation indices. Our data show additive or synergistic effects ofthese combinations in the majority of cell lines tested [Table 1(CC-401), Supplementary Table S3 (SP600125)], implying theimportance of JNKpathway in drug resistance. However, not all ofthe effects are hypoxia specific in this setting. For example, thecombination of CC-401 with oxaliplatin demonstrates highersynergy under hypoxia in HT29, SW620, HCT116, HCT15, andLoVo cell lines (5 out of 6), with median CI of 0.81 and 0.48 foroxic and hypoxic conditions, respectively. Enhanced cytotoxicityof SN-38 is further significantly augmented by hypoxia only inHT29 cells, with median CI values for whole panel of 0.56 fornormoxia and 0.54 in hypoxic cells. The cytotoxic interactionbetween CC-401 and 5-FU is also synergistic in oxic conditions(median CI ¼ 0.88) and hypoxia (median CI ¼ 0.64). Althoughhigher synergism under hypoxia is observed in HT29, HCT116,and KM12 cells, we further narrowed the field of our study tooxaliplatin, because neither HCT116 nor KM12 demonstratesincreased resistance to 5-FU under hypoxia (Fig. 1B). Combina-tions of the drugs with SP600125, while also demonstratingrather modest hypoxic specificity (Supplementary Table S3), stillare mostly synergistic. Thus, inhibition of JNK signaling by CC-401 enhances cytotoxicity of chemotherapy across the board inboth normal and hypoxic conditions. To concentrate further onthe JNK involvement in cytotoxicity of chemotherapy underhypoxia, we focused on cell models that demonstrate both anincrease in oxaliplatin resistance under hypoxia and activation ofthe JNK pathway (HT29 and SW620-responsive lines), and a cellline that demonstrates neither (HCT116-nonresponsive line) as acontrol. Oxaliplatin or SN-38 alone in normal conditions inducesJNK signaling only in HCT116 cells, whereas in two others it iseither not activated (HT29) or not altered (SW620), indicatingthat in these lines JNK induction is a consequence of hypoxia(Supplementary Fig. S1). We then investigated which cellularresponses to oxaliplatin under hypoxia are mediated by the JNKpathway beginning with an assessment of DNA damage in thissetting.
Hypoxia induces DNA damage response in cell-specificmannerPlatinum compounds result in DNA adduct formation, which
in turn initiate several cellular responses. Involvement of JNK inthe DNA damage response, by controlling directly or indirectlyDNA damage repair and cell death, among other processes, was
2,000
1,800
1,600
1,400
1,200
1,000
800
600
400
200
01 3 6 9 12 15 18 21 24 27 30 33 36
Days after treatment
Tum
or
volu
me
(1×1
03 m
m)
Control
CC-401
Bev
Bev+CC-401
L-OHP
L-OHP+CC-401
Bev+L-OHP
Bev+L-OHP+CC-401
Figure 3.JNK inhibition by CC-401 sensitizes mouse colon cancer xenografts tooxaliplatin and bevacizumab. Graph shows volumes of HT29-derived tumorsin mice treated as described in Materials and Methods. � , P < 0.05.
Table 2. Effect of CC-401 on the growth delay of HT29-derived tumors treatedwith bevacizumab, oxaliplatin, and the combination
NOTE: For tumor growth delay calculations, mice continued to receive treat-ments until the tumor volume reached 800 mm3.aTumor growthdelaywas calculated as time (days) needed for treated tumors togrow from 200 to 800mm3minus the time needed for control tumor to grow tothe same size.
JNK Inhibition Potentiates Therapy in Colon Cancer Models
www.aacrjournals.org Clin Cancer Res; 21(18) September 15, 2015 4147
shown in multiple models (27). We therefore asked whether JNKinhibition by CC-401 would modify DNA damage responsesinduced by hypoxia, oxaliplatin, or both in HT29, SW620, andHCT116 cell lines. We showed that 1�IC50 concentrations of CC-401 (3, 6.5 and 3.5 mmol/L for HT29, SW620 and HCT116,respectively) are sufficient to inhibit hypoxic JNK signaling andused them in all further experiments (Fig. 2A and B). For assess-ment ofDNAdamage, cellswere subjected tohypoxia for 24hourswith or without oxaliplatin and the levels of phosphorylated p-H2AX measured by FACS analysis. We used oxaliplatin at con-centrations of 0.7, 0.4, and 0.9 mmol/L, for HT29, SW620, andHCT116 cells, respectively (1xIC50 for oxic conditions).We found(Fig. 2C) that hypoxia causes a significant increase in p-HA2X inall cell lines (4–5 fold, comparedwith untreated control), but JNKinhibition by CC-401 (one hour treatment before hypoxia) aug-ments it further only in HT29 and SW620 cell lines. This obser-vation serves to emphasize that in models in which JNK isactivated by hypoxia, its inhibition may have therapeutic con-sequences. We also evaluated activation of Chk1 and Chk2kinases at 24 hours of hypoxia, which was brisk with or withoutCC-401 (Fig. 2A and B). Our data do not permit a firm quanti-tative conclusion, but suggest that there is no inhibition of DNAdamage signaling by CC-401. Together these results suggest thatCC-401 may exert its synergistic activity by increasing DNAdamage in "responsive" cell lines.
JNK inhibition by CC-401 sensitizes mouse colon cancerxenografts to oxaliplatin and bevacizumab
To determine the effect of JNK inhibition on therapy withbevacizumab, oxaliplatin, and the combination in vivo, micebearingHT29 xenograftswere randomly divided into eight groupsfor treatments as follows: PBS-treated control, CC-401 alone,bevacizumab alone, bevacizumab and CC-401 in combination,oxaliplatin alone, oxaliplatin and CC-401 in combination, bev-acizumab and oxaliplatin in combination, and bevacizumab,oxaliplatin and CC-401 in combination. Figure 3 shows effectsof these treatments on growth of the tumors. The tumor growthdelay for 2 days was observed for CC-401 alone (Table 2). Amuchgreater impact was seen with the combinations of CC-401 withoxaliplatin alone (6 vs. 18 days), bevacizumab alone (7 vs. 9days), and with the combination of both (11 vs. 19 days).Therefore in all treatments, the strategy of incorporating CC-401 increased the tumor growth delay significantly, suggestingthat JNK-signaling inhibition may have an impact on efficacy ofboth oxaliplatin and antiangiogenesis therapy (Table 2).
Tumors from mice treated as described above were processedfor immunostaining of phospho-JNK and phospho-cJun proteinsto assess JNK signaling in xenografts and effects of CC-401. IHCanalysis revealed the cytoplasmic localization of p-JNK in control,bevacizumab-, and oxaliplatin-treated tumors (Fig. 4A), whereasp-cJun staining was nuclear in all samples (Fig. 4B). The stainingof p-JNKwasmoderately induced in bevazicumab andoxaliplatintreatments as compared with control, and in the CC-401–treatedsamples, p-cJun content was significantly lower, consistent witheffective JNK inhibition. To identify additional effects of com-
bined treatments on HT29-derived xenografts, tumors were alsostained for blood vessels, cell viability, and DNA damage usingCD31, Ki-67, and p-H2AX antibodies. As evidenced in Fig. 4C,bevacizumab inhibited blood vessels density in the treated sam-ples, with CC-401 and oxaliplatin exerting no effect on its anti-angiogenic properties (Fig. 4C). Staining with cell Ki-67 revealedthe lowest proliferation when bevacizumab, oxaliplatin, and CC-401were combined (Fig. 4D). Finally,DNAdamagewasmodestlyelevated in combined treatments with CC-401 (Fig. 4E). Thus, thein vivo results were in accordance with in vitro studies in HT29colon adenocarcinoma. We then sought to determine whetherinhibition of JNK isoforms could affect oxaliplatin cytotoxicity inthis cell line.
Downregulation of JNK1 sensitizes HT29 cells to oxaliplatinunder hypoxia
For these studies, we isolated HT29-derived cell lines stablyexpressing dominant-negative, non-phosphorylatablemutants ofeither JNK1 (HTJ1.3) or JNK2 (HTJ2.2), with a cell line transfectedwith empty vector as control (HTLX). Our results show that underhypoxic exposure, phosphorylation of c-Jun is diminished in thecell linewith impaired JNK2,while the protein level is comparablewith that of parental cells. On the other hand, inHTJ1.3 cells bothexpression and phosphorylation levels of c-Jun are elevated (Fig.5A). We assessed sensitivity of these HT29 derivatives to oxali-platin, SN-38, and 5-FU, and found that JNK1 deficiency abro-gates hypoxic resistance to oxaliplatin, whereas downregulationof JNK2 slightly increases resistance when compared with controlcells (hypoxic IC50 of 0.8, 2.9, and 2.3 mmol/L, respectively). Theeffects of JNK modulation were less pronounced in hypoxic cellstreated with SN38 (27, 25, and 34 nmol/L for HTLX, HTJ1.3 andHTJ2.2 cells) anddownregulationofboth JNK isoforms sensitizedhypoxic HT29 to 5-FU (11, 8, and 5.5 mmol/L, respectively; Fig.5B). These data suggest differing inputs of JNK isoforms in thesensitivity of colon cancer cell lines to chemotherapeutic drugs,and warrant further investigation to identify more selectiveapproaches to the pathway.
DiscussionThe JNK signaling pathway regulates a multitude of cellular
processes involved in both proliferation and cell death, and itsderegulation (most often activation) has been implicated inseveral diseases including neurologic and cardiac disorders,inflammation, and cancer (28). In many of these diseases,inhibition of overactivated JNK may be beneficial. In cancer,the effects of JNK inhibition, both pharmacologically and ingenetically modified models, are opposing in many tumors,reflecting the interaction of the JNK pathway with molecularcharacteristics of the cancer and/or host. Activation of the JNKpathway in colonic tissues and biopsies was observed ininflammatory bowel disease and Crohn's disease (28), but incolon cancer, it is less well described. A contribution of JNK tocarcinogenesis by cooperation of JNK and b-catenin pathwaysin activation of c-Jun was documented in APCmin mice (29).
Figure 4.IHC analysis of HT29-derivedmouse xenografts confirms effects of JNK inhibition on cytotoxicity of oxaliplatin under hypoxia. Tumorswere assessed for localizationand phosphorylation status of JNK (A) and c-Jun (B), for blood vessel density (C), viability (D), and DNA damage (E). Corresponding graphs represent intensity ofimmunostaining, quantified as described in Materials and Methods, in tumors treated with (striped bars) or without (black bars) CC-401.
JNK Inhibition Potentiates Therapy in Colon Cancer Models
www.aacrjournals.org Clin Cancer Res; 21(18) September 15, 2015 4149
More studies revealed that JNK/b-catenin pathways cooperateat a transcriptional level through c-Jun/TCF4 common targetgenes activated in cancer, such as c-jun itself, c-myc, cyclin D1,MMP7, and others (30). In chemically induced mouse livercancer, JNK deficiency also leads to impaired proliferation,through an increase in the cell-cycle inhibitor p21, and reducedexpression of c-Myc (31). Reduction in tumor formationand growth after JNK inhibition was observed in intestinal,lung, and ovarian cancer models, among others (32–34). Onthe other hand, JNK deficiency resulted in increased tumorformation in models of breast and hepatocellular cancer (35,36), and in a genetically engineered mouse model of pancreaticcancer (37). In our study, JNK was variably activated bothconstitutively and, even more so, in response to hypoxia incolon cancer cells (Fig. 1A). However, as with many therapeuticapproaches, optimal translation of these findings to the clinicmust recognize the potential for a deleterious effect in sometumors, and perhaps limit the duration of pathway inhibitionto the period surrounding cytotoxic drug administration.
Our analysis of 12 cell lines suggests that JNK activationcontributes to the survival of colon cancer cells, since its inhibitionwith CC-401 resulted in higher sensitivity to oxaliplatin, SN-38,and 5FU in both normal and hypoxic conditions (Table 1). In amajority of cell lines, an increase in chemotherapy resistanceunder hypoxia was associated with JNK induction (Fig. 1B). Celllines in which that was not the case (DLD1, HCT15, HCT116 and
its derivative, HCT116p53�/�) all have the mismatch repairdeficiency genotype, representing a relatively small proportionof patients with metastatic colon cancer, which can differ frommost common types in their behavior (38). Our data suggest thatJNK could be a valid target to enhance efficacy of chemotherapy incolon tumors.
The outcome of JNK inhibition in cell lines also varied with theDNA-damaging stimulus involved, especially under hypoxic con-ditions. Although it was almost uniformly additive or synergisticin all cell lines, significant hypoxia-specific enhancement of SN-38and 5-FU toxicity by CC-401 was shown only in one (HT29) andthree (HT29, HCT116, and KM12) cell lines, out of six, respec-tively. By far, the best hypoxic interactionwith JNK inhibitionwasobserved for oxaliplatin: five cell lines demonstrate clear hypoxia-specific synergism of these compounds in combination (Table 1).The basis for difference is, most likely, the variability of cellularresponses to DNA damage (39) induced by particular chemo-therapy, which in turn depend on genetic makeup of the cells. Inour panel, synergism of oxaliplatin in the combination with CC-401 is seemingly based on DNA damage extent and ensuingenhancement of cell death, either by apoptosis or necrosis (40,41). In addition, the inhibition of JNK by CC-401 could alsoattenuate induction of protective hypoxia-induced autophagy(26) in some cell lines, but this needs to be confirmed further.On the basis of the data presented here, the increase in DNAdamage consequent upon JNK inhibition may be the most
0 3 6 24 0 3 6 24 0 3 6 24
JNK2
JNK1
HA-tag
Actin
HTLX HTJ1.3 HTJ2.2
p-cJun
c-Jun
dnJNK2dnJNK1
Hypoxia
0
0.5
1
1.5
2
2.5
3
3.5
HTJ2.2HTJ1.3HT.LX
Oxa
lipla
tin
IC50
(mm
ol/L
) OxicHypoxic
0
2
4
6
8
10
12
14
HTJ2.2HTJ1.3HTLX
5-F
U IC
50 (
mmo
l/L)
OxicHypoxic
0
5
10
15
20
25
30
35
40
HTJ2.2HTJ1.3HTLX
SN
-38
IC50
(n
mo
l/L)
OxicHypoxic
SN38 5-FUOxaliplatin
A
B
Figure 5.Downregulation of JNK isoforms results in varying effects on cytotoxicity of chemotherapeutic drugs: inhibition of JNK1 sensitizes HT29 cells to oxaliplatin underhypoxia. A, HT29-derived monoclonal cell lines with downregulated JNK1 or JNK2 were subjected to Western blot analysis to assess hypoxia-inducedsignaling through JNK. B, graphs depict IC50 concentrations of oxaliplatin, SN-38, and 5-FU in HTLX, HTJ1.3, and HTJ2.2 cell lines in normal (black bars) or hypoxic(striped bars) conditions, based on the results of MTT assays.
Vasilevskaya et al.
Clin Cancer Res; 21(18) September 15, 2015 Clinical Cancer Research4150
important driver of oxaliplatin sensitization in hypoxic coloncancer cell lines.
Combinations of bevacizumab, theVEGF-bindingmonoclonalantibody, with oxaliplatin, irinotecan, and 5-FU are effective incolorectal cancer treatment (42). By inhibiting the tumor vascu-lature, bevacizumab and other antiangiogenic agents cause sub-stantial alterations in the tumormicroenvironment (43).We havepreviously demonstrated that bevacizumab treatment induceshypoxia in colon cancer xenografts, but tumor shrinkage as aconsequence of treatment is variable (25, 44) and depends onsusceptibility to hypoxia-induced cell death (25). In this study,the growth of HT29-derived xenografts showed noticeable delaywhen mice were treated with bevacizumab or oxaliplatin alone(Fig. 3), but the best results were achieved in triple treatment,when oxaliplatin and CC-401 were combined in a bevacizumab-induced hypoxic environment (Table 2; Fig. 4C). These datasupport a positive impact of JNK inhibition on cytotoxicity ofoxaliplatin in hypoxic tumors, and support clinical testing of thecombination.
The complexity of cellular responses to JNKpathway inhibitionin varying contexts led to interest in development of betterpharmacologic JNK inhibitors, including those against specificisoforms (45, 46). These inhibitors may facilitate an understand-ing of the distinctive, and sometime opposing, functions of JNKisoforms (47, 48). At this point, however, a specific small-mol-ecule inhibitor of JNK1 or JNK2 is not readily available. Toinvestigate the effect of JNK isoform inhibition on cytotoxicityof the chemotherapy, we performed initial experiments usingdominant-negative constructs of JNK1 and JNK2 in HT29 cells.This approach has the potential to lower the ability of JNKs tosubstitute for each other, since nonphosphorylatable proteinswould compete with corresponding endogenous kinase for bind-ing both to upstream activators and to downstream targets. Thefact that inhibition of either JNK isoform does not significantlyalter sensitivity to SN38 under hypoxia, and leads to increasedsensitivity to 5-FU, suggests differing input of JNK isoforms insignal transduction induced by each drug, and points to thebenefit of pan-JNK inhibition used alongside 5-FU treatment. Inthe case of oxaliplatin, however, there is an obvious difference incellular responses: inhibition of JNK1 in HT29 completelyabolishes hypoxia-induced resistance to the drug, whereas inhi-
bition of JNK2 results in a slightly opposite effect. This resultsuggests a prosurvival function for JNK1 under these circum-stances, which could include induction of protective autophagy,DNA damage repair, or other cellular pathways. Although in ourpreliminary experiments, inhibition of JNK2 led to a slightincrease in resistance to oxaliplatin under hypoxia, diminishedexpression, and activation of c-Jun (Fig. 5A) could nevertheless beof a benefit considering the pro-oncogenic features of this proteinas a part of AP-1. Our data suggest that inhibition of JNK1 wouldbe more advantageous when combined with oxaliplatin fortreatment of solid tumors, but unless specific inhibitors for eachisoform is available, pan-inhibition of JNK would work as well,because there is little downside form JNK2 inhibition.
Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.
Authors' ContributionsConception and design: I.A. Vasilevskaya, M. Selvakumaran, S.R. Goldstein,J.D. Winkler, P.J. O'DwyerDevelopment of methodology: I.A. Vasilevskaya, M. Selvakumaran,L. Cabal Hierro, S.R. Goldstein, J.D. Winkler, P.J. O'DwyerAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): I.A. Vasilevskaya, M. Selvakumaran, L. Cabal Hierro,P.J. O'DwyerAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): I.A. Vasilevskaya, M. Selvakumaran, L. Cabal Hierro,P.J. O'DwyerWriting, review, and/or revision of the manuscript: I.A. Vasilevskaya,M. Selvakumaran, L. Cabal Hierro, S.R. Goldstein, J.D. Winkler, P.J. O'DwyerAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): L. Cabal Hierro, S.R. Goldstein, P.J. O'DwyerStudy supervision: P.J. O'Dwyer
Grant SupportThis work was supported in part by R01CA139003 from NCI, NIH.The costs of publication of this articlewere defrayed inpart by the payment of
page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received February 12, 2015; revised May 8, 2015; accepted May 12, 2015;published OnlineFirst May 28, 2015.
References1. Vaupel P, Mayer A. Hypoxia in cancer: significance and impact on clinical
outcome. Cancer Metastasis Rev 2007;26:225–39.2. Brown JM, Giaccia AJ. The unique physiology of solid tumors: oppor-
tunities (and problems) for cancer therapy. Cancer Res 1998;58:1408–16.
3. Rowher N, Cramer T. Hypoxia-mediated drug resistance: Novel insights onthe functional interaction of HIFs and cell death pathways. Drug ResUpdates 2011;14:191–201.
5. Philip B, Ito K,Moreno-S�anchez R, Ralph SJ. HIF expression and the role ofhypoxic microenvironments within primary tumours as protective sitesdriving cancer stem cell renewal and metastatic progression. Carcinogen-esis 2013;34:1699–707.
6. Ghattass K, Assah R, El-SabbanM, Gali-Muhtasib H. Targeting hypoxia forsensitization of tumors to radio- and chemotherapy. Curr Cancer DrugTargets 2013;13:670–85.
7. SunQ, Li X. Targeting cyclic hypoxia to prevent malignant progression andtherapeutic resistance of cancers. Histol Histopathol 2015;30:51–60
8. Cummins EP, Taylor CT.Hypoxia-responsive transcription factors. PflugersArch 2005;450:363–71.
9. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat RevCancer 2003;3:859–68.
10. Chang L, Karin M. Mammalian MAP kinase signaling cascades. Nature2001;410:37–40.
11. Fanger GR, Gerwins P, Widmann C, Jarpe MB, Johnson GL. MEKKs, GSKs,PAKs and Tpls: upstream regulators of the c-Jun amino-terminal kinases?Curr Opinion Gen Develop 1997;7:67–74.
12. Zheng CF, Guan KL. Properties of MEKs, the kinases that phosphorylateand activate the extracellular signal-regulated kinases. J Biol Chem1993;268:23933–9.
13. Enslen H, Raingeaud J, Davis RJ. Selective activation of p38 mitogen-activated protein (MAP) kinase isoforms by the MAP kinase kinases MKK3and MKK6. J Biol Chem 1998;273:1741–8.
14. Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell2000;103:239–52.
15. Deschenes-Simard X, Kottakis F, Meloche S, Ferbeyre G. ERKs in cancer:friends or foes? Cancer Res 2014;74:412–9.
www.aacrjournals.org Clin Cancer Res; 21(18) September 15, 2015 4151
JNK Inhibition Potentiates Therapy in Colon Cancer Models
16. Koul HK, Pal M, Koul S. Role of p38 MAP kinase signal transduction insolid tumors. Genes Cancer 2013;4:342–59.
17. Tournier C. The 2 faces of JNK signaling in cancer. Genes Cancer2013;4:397–400.
18. Yao K-S, Xanthoudakis S, Curran T,O'Dwyer PJ. Activation of AP-1 and of anuclear redox factor, Ref-1, in the response of HT29 colon cancer cells tohypoxia. Mol Cell Biol 1994;14:5997–6003.
19. O'Dwyer PJ, Yao K-S, Ford P, Godwin AK, Clayton M. Effects of hypoxiaupon detoxicating enzyme activity and expression in HT29 colon adeno-carcinoma cells. Cancer Res 1994;54:3082–7.
20. Vasilevskaya IA,O'Dwyer P J. Effects of geldanamycin on signaling throughactivator-protein 1 in hypoxic HT29 human colon adenocarcinoma cells.Cancer Res 1999;59:3935–40.
21. Vasilevskaya IA, Selvakumaran M, O'Dwyer PJ. Disruption of signalingthrough SEK1 and MKK7 yields differential responses in hypoxic coloncancer cells treated with oxaliplatin. Mol Pharmacol 2008;74:246–54.
22. Vasilevskaya IA, O'Dwyer PJ. Role of Jun and Jun kinase in resistance ofcancer cell to therapy. Drug Res Updates 2003;6:147–56.
23. Chou T-C. Theoretical basis, experimental design and computerized sim-ulation of synergism and antagonism in drug combination studies. Phar-macol Rev 2006;58:621–81.
24. Chou T-C, Martin N. CompuSyn software for drug combinations and forgeneral dose-effect analysis, and user's guide. ComboSyn, Inc. Paramus, NJ2007. [www.combosyn.com].
25. Selvakumaran M, Yao KS, Feldman MD, O'Dwyer PJ. Antitumor effect ofthe angiogenesis inhibitor bevacizumab is dependent on susceptibility oftumors to hypoxia-induced apoptosis. Biochem Pharmacol 2008;75:627–38.
26. Selvakumaran M, Amaravadi RK, Vasilevskaya IA, O'Dwyer PJ. Autophagyinhibition sensitizes colon cancer cells to antiangiogenic and cytotoxictherapy. Clin Cancer Res 2013;19:2995–3007.
27. Picco V, Pages G. Linking JNK activity to the DNA damage response. GenesCancer 2013;4:360–8.
28. Sabapathy K. Role of JNK pathway in humandiseases. ProgMol Biol TranslSci 2012;16:145–69.
29. Nateri AS, Spenser-Dene B, Behrens A. Interaction of phosphorylated c-Junwith TCF4 regulates intestinal cancer development. Nature 2005;437:281–5.
30. Saadeddin A, Babaei-Jadidi R, Spenser-Dene B, Nateri AS. The linksbetween transcription, b-catenin/JNK signaling, and carcinog�enesis. MolCancer Res 2009;7:1189–96.
32. Fujishita T, Aoki M, Taketo MM. JNK signaling promotes intestinal tumor-igenesis through activation of mTOR complex 1 in Apc(D716) mice.Gastroenterology 2011;140:1556–63.
33. Khatlani TS, Wislez M, Sun M, Srinivas H, Iwanaga K, Ma L, et al. c-Jun N-terminal kinase is activated in non-small-cell lung cancer and promotesneoplastic transformation in human bronchial epithelial cells. Oncogene2007;26:2658–66.
34. Vivas-Mejia P, Benito JM, Fernandez A, Han HD, Mangala L, Rodriguez-Aguayo C, et al. c-JUN-NH2-kinase-1 inhibition leads to antitumor activityin ovarian cancer. Clin Cancer Res 2010;16:184–94.
35. Cellurale C, Girnius N, Jiang F, Cavanagh-Kyros J, Lu S, Garlick DS, et al.Role of JNK inmammary gland development and breast cancer. Cancer Res2012;72:472–81.
36. Das M, Garlick DS, Greiner DL, Davis RJ. The role of JNK in the develop-ment of hepatocellular carcinoma. Genes Dev 2011;25:634–45.
37. Davies CC, Harvey E, McMahon RF, Finegan KG, Connor F, Davis RJ, et al.Impaired JNK signaling cooperates with KrasG12D expression to acceleratepancreatic ductal adenocarcinoma. Cancer Res 2014;74:3344–56.
38. Wilding JL, Bodmer WF. Cancer cell lines for drug discovery and devel-opment. Cancer Res 2014;74:2377–84.
39. Surova O, Zhivotovsky B. Various models of cell death induced by DNAdamage. Oncogene 2013;32:3789–97.
40. Hayward R, Macpherson JS, Cummings J, Monia BP, Smyth JF, Jodrell DI.Enhanced oxaliplatin-induced apoptosis following antisense Bcl-xl down-regulation is p53 and Bax dependent: genetic evidence for specificity of theantisense effect. Mol Cancer Ther 2004;3:169–78.
41. Rakitina TV, Vasilevskaya IA, O'Dwyer PJ. Inhibition of G1/S transitionpotentiates oxaliplatin-induced cell death in colon cancer cell lines. Bio-chem Pharmacol 2007;73:1715–26.
42. Kirstein MM, Lange A, Prenzler A, Manns MP, Kubicka S, Vogel A. Targetedtherapies in metastatic colorectal cancer: a systematic review and assess-ment of currently available data. Oncologist 2014;19:1156–68.
43. Willett CG, Boucher Y, di Tomaso E, Duda DG, Munn LL, Tong RT, et al.Direct evidence that the VEGF-specific antibody bevacizumab has anti-vascular effects in human rectal cancer. Nat Med 2004;10:145–7.
44. Yao K, Gietema JA, Shida S, SelvakumaranM, Fonrose X, Haas NB, et al. Invitrohypoxia-conditioned colon cancer cell lines derived fromHCT116 andHT29 exhibit altered apoptosis susceptibility and amore angiogenic profilein vivo. Br J Cancer 2005;93:1356–63.
45. Bubici C, Papa S. JNK signaling in cancer: in need of new, smartertherapeutic targets. Br J Pharm 2014;171:24–37.
46. Messoussi A, Feneyrolles C, Bros A, Deroide A, Dayde-Cazals B, Cheve G,et al. Recent progress in the design, study, and development of c-Jun N-terminal kinase inhibitors as anticancer agents. Chem Biol 2014;11:1433–43.
47. Bogoyevitch MA. The isoform-specific functions of the c-Jun-terminalkinases (JNKs): differences revealed by gene targeting. Bioassays2006;28:923–34.
48. Bode AM, Dong Z. The functional contrariety of JNK. Mol Carcinog2007;46:591–8.
Clin Cancer Res; 21(18) September 15, 2015 Clinical Cancer Research4152
2015;21:4143-4152. Published OnlineFirst May 28, 2015.Clin Cancer Res Irina A. Vasilevskaya, Muthu Selvakumaran, Lucia Cabal Hierro, et al. DNA-Damaging AgentsInhibition of JNK Sensitizes Hypoxic Colon Cancer Cells to
Updated version
10.1158/1078-0432.CCR-15-0352doi:
Access the most recent version of this article at:
