Combined Effects of Suberoylanilide Hydroxamic Acid and ......Combined Effects of Suberoylanilide Hydroxamic Acid and Cisplatin on Radiation Sensitivity and Cancer Cell Invasion in

Small Molecule Therapeutics

Combined Effects of Suberoylanilide HydroxamicAcid and Cisplatin on Radiation Sensitivityand Cancer Cell Invasion in Non–SmallCell Lung CancerJianguo Feng1, Shirong Zhang2,3, Kan Wu3,4, Bing Wang3,4, Jeffrey Y.C.Wong5,Hong Jiang2, Rujun Xu2, Lisha Ying1, Haixiu Huang2, Xiaoliang Zheng6,Xufeng Chen7, and Shenglin Ma2,3

Abstract

Lung cancer is a leading cause of cancer-related mortalityworldwide, and concurrent chemoradiotherapy has beenexplored as a therapeutic option. However, the chemothera-peutic agents cannot be administered for most patients at fulldoses safely with radical doses of thoracic radiation, and furtheroptimizations of the chemotherapy regimen to be given withradiation are needed. In this study, we examined the effects ofsuberoylanilide hydroxamic acid (SAHA) and cisplatin on DNAdamage repairs, and determined the combination effects ofSAHA and cisplatin on human non–small cell lung cancer(NSCLC) cells in response to treatment of ionizing radiation(IR), and on tumor growth of lung cancer H460 xenograftsreceiving radiotherapy. We also investigated the potential dif-ferentiation effect of SAHA and its consequences on cancer cell

invasion. Our results showed that SAHA and cisplatin com-promise distinct DNA damage repair pathways, and treatmentwith SAHA enhanced synergistic radiosensitization effects ofcisplatin in established NSCLC cell lines in a p53-independentmanner, and decreased the DNA damage repair capability incisplatin-treated primary NSCLC tumor tissues in response toIR. SAHA combined with cisplatin also significantly increasedinhibitory effect of radiotherapy on tumor growth in the mousexenograft model. In addition, SAHA can induce differentiationin stem cell–like cancer cell population, reduce tumorigenicity,and decrease invasiveness of human lung cancer cells. In con-clusion, our data suggest a potential clinical impact for SAHAas a radiosensitizer and as a part of a chemoradiotherapyregimen for NSCLC. Mol Cancer Ther; 15(5); 1–12. �2016 AACR.

IntroductionLung cancer is themost commoncancer and the leading cause for

cancer-relatedmortalityworldwide (1). Themajorityof lung canceris NSCLS (non–small cell lung cancer) that includes squamous cellcarcinoma, adenocarcinoma, and large cell carcinoma.One thirdofthese patients are diagnosed with stage III disease when curative

treatment is extremely limited (2). Despite the tremendous effortsand progress in lung cancer research, and the use of aggressivemultimodal chemo- and radiotherapy, the overall treatment out-come for these NSCLC patients remains poor.

For patients with advanced stages of NSCLC when surgicalexcision is not an option, the combined treatment of radiotherapyand chemotherapy is extensively used. The first-line therapytypically includes a platinum derivative, such as cis-diaminedi-chloroplatinum (II; cisplatin; refs. 3, 4). Meta-analyses based onindividual data from randomized clinical trials have demonstrat-ed statistically significant benefits of combining chemotherapywith radiation for stage III NSCLC patients, and the greatestdifference in efficacy was observed in the cisplatin-based trials(5). In addition, meta-analysis data also showed that concurrentchemotherapy and radiation (ChRT) significantly improved sur-vival when compared with sequential chemotherapy and radia-tion, and most effective treatment observed was the immediateconcurrent ChRT (5, 6).

However, most commonly used chemotherapy regimens can-not be administered at full doses safely with radical doses ofthoracic radiation, and a conclusion arising from these efforts isthat concurrent ChRT produces the highest cure rates but with anincreased level of toxicity (7, 8). Notable, data from clinical trialswith cisplatin-based concurrent ChRT showed that scheduleswithlow-dose chemotherapy might reduce the incidence and level ofside toxicities, including acute esophagitis and hematologic toxi-cities (8). It is therefore possible to hypothesize that further

1Cancer Research Institute and Key Laboratory Diagnoses and Treat-ment Technology on Thoracic Oncology, Zhejiang Cancer Hospital,Hangzhou, China. 2Department of Oncology, Affiliated HangzhouHospital of Nanjing Medical University, Hangzhou, China. 3AffiliatedHangzhou First People's Hospital of Zhejiang Chinese Medical Uni-versity, Hangzhou, China. 4Department of Radiation Oncology, Hang-zhou Cancer Hospital, Hangzhou, China. 5Department of RadiationOncology, City of Hope Cancer Center, Duarte, California. 6Centre ofMolecular Medicine, Zhejiang Academy of Medical Sciences, Hang-zhou, China. 7Department of Pathology and Laboratory Medicine,University of California at Los Angeles, Los Angeles, California.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

J. Feng, S. Zhang, and K. Wu contributed equally to this article.

Corresponding Author: Shenglin Ma, Affiliated Hangzhou Hospital of NanjingMedical University, 261 Huansha Road, Hangzhou, Zhejiang 310022, China.Phone: 86-571-5600-7908; Fax: 86-571-8791-4773; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-15-0445

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org OF1

on April 6, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst February 2, 2016; DOI: 10.1158/1535-7163.MCT-15-0445

http://mct.aacrjournals.org/

optimization of chemotherapy regimen to be givenwith radiationmay help for better treatment outcomes, with reduced toxicity.

In chemoradiotherapy strategies for cancer patients, the che-motherapy aim tominimize the risk of distant metastasis and theradiotherapy will preserve locoregional control (9). A chemother-apeutic drug may also act as a radiosensitizer and increase theeffect of radiation treatment (10). Of note, studies have shownthat cisplatin can increase radiosensitization (11). However,studies also revealed that cisplatin treatment is associated withsignificant toxic effects and requires fluid hydration, which can beproblematic in patients with cardiovascular disease. In addition,prolonged cisplatin treatment may promote the emergence ofresistant tumors (12).

In this study, we investigated the synergistic radiosensitizationeffects of Suberoylanilide Hydroxamic Acid (SAHA) and cisplatinon human lung cancer cells. We demonstrate here that SAHA canpromote the radiosensitization effects of cisplatin on human lungcancer, which may be mediated through interaction of impairednon-homologous end-joining (NHEJ or EJ) and homology-directed repair (HDR or DR) for repair of DNA double-strandbreak (DSB) caused by treatment of ionizing radiation (IR). Ourdata also showed that SAHA can induce differentiation of cancerstem cells (CSC) and reduce the invasiveness of lung cancer cells.

Materials and MethodsReagents

SAHA and cisplatin were purchased from Sigma-Aldrich. Anti–DNA-pKcs and anti–phospho-histone H2A.X (ser-139) antibo-dies were from Upstate Biotechnology. Anti-RAD51, anti-KU70,anti-syndecan 3, anti-involucrin, and anti-p53 antibodies werefrom Santa Cruz Biotechnology. Anti–b-actin antibody was fromCell Signaling Technology. siRNA oligos for RAD51, p53 andDNA-PKcs, and control siRNA-A were also from Santa CruzBiotechnology. Plasmids pimEJ5GFP (http://www.addgene.org/44026) and pDRGFP (http://www.addgene.org/26475) wereprovided byDr. Jeremy Stark (City of Hope, Duarte, CA). PlasmidpcDNA3.1-C was from Thermo Fisher Scientific. PlasmidpcDNA3.1-p53wt has been described previously (13). EnzymesI-SceI and EcoR V were from New England Biolabs.

Cell cultureHuman NSCLC cell lines H460, A549, H1299, and MCF-10A

were obtained directly from the ATCC, which has providedcertification of authentication as published on their website(i.e., using Karyotype and STR Profiling analyses). A549 cells weregrown in F12-K medium supplemented with 10% FBS (Omegascientific). H460 and H1299 cells were grown in RPMI-1640medium supplementedwith 10%FBS.MCF-10A cells were grownin MEGM Mammary Epithelial Cell Growth Medium (Lonza).U2OS reporter cell lines, which consist of theU2OSosteosarcomacell line with integrated copies of the EJs-GFP, DR-GFP, and SSA-GFP reporters along with a unique I-SceI cleavage site, have beendescribed previously (14, 15), and were maintained in DMEMmedium with 2 mmol/L L-Glutamine, 4.5 g/L glucose, and 10%FBS. Hamster lung fibroblast cell lines V79 (wild-type) and VC8(defective in the BRCA2 gene), andChinese hamster ovary (CHO)cell lines AA8 (wild-type) and V3 (defective in DNA-PKcs expres-sion)weremaintained inDMEMmediumas described previously(14). All cell lines were tested as negative Mycoplasma contam-ination with Cell Culture Contamination Detection Kit (Thermo

Fisher Scientific), and none of the cell lines was cultured longerthan 6 months.

Stable transfectionTransfections were performed with electroporation with Gene

Pulse Xcell (Bio-Rad) as per the manufacturer's instructions.Stable transfectants were selected by G418 (Sigma).

Clonogenic survival assaysLog-phase cells were treated with SAHA for 72 hours or with

cisplatin for 24hours. Cellswere trypsinized andplated for colonyformation. Ionizing irradiation (IR) was delivered immediatelyafter cell plating. When combination treatment was applied, cellswere pretreated with SAHA for 72 hours, and cisplatin was addedinto cell cultures 24 hours before trypsinization. DMSO wasincluded as control. Survived colonies were stained with crystalviolet, and colonies consisting >50 cells were directly scored.Average numbers for survival colonies were plotted versus dosesof SAHA, cisplatin or IR to determine the lethal doses (LD), orsurvival fractions.

In experiments with siRNA, cells were transiently transfectedwith siRNAoligos. Twenty-four hours later, cells were then treatedwith SAHA, cisplatin, and/or IR as described above.

Median effect analysisMedian effect analysis was performed as described previously

(16). Briefly, cells were treated with increasing total doses ofcisplatin, SAHA, and IR with constant ratio of doses based onthe corresponding LD50, and cell survival was determined usingclonogenic survival assay. A plot of the log of the total dose versuslog of the reciprocal of the fraction of cells affected minus 1yielded linear plot. The slope and y-intercept from these plotswereused to calculate the CI (17, 18).

Immunofluorescence analysisAfter treatments, cells were washed twice with Caþþ/Mgþþ-free

PBS, and fixed in 4% paraformadehyde. Immunofluorescenceanalyses for g-H2A.X were performed as previously reported(14). Images were acquired with LSM 510 confocal microscope(Zeiss) with�40 objective and processed by Photoshop (Adobe).At least 100 cells from each experiment were selected at randomand were counted to calculate the percentage of cells as "positive"for g-H2A.X foci if they displayed >5 discrete dots in nuclei.

DNA damage repair assaysA total of 1 � 105 U2OS reporter cells were plated in 12-well

plate. Twenty-four hours later, cells were transfected with 0.8 mgpCBASce by using Lipofectamine 2000 transfection reagent (LifeTechnologies), and 1 mmol/L SAHA or 2 mmol/L cisplatin wereadded into culturemedium 3 hours later after the initiation of thetransfection. After exposure for 24 hours, cisplatin was removedfrom cell culture, and cells were maintained in fresh medium forcisplatin treatment or in SAHA-containing medium until analysisas described previously (14). GFP-positive cells were quantifiedby flow cytometric analysis. Up to 5 � 104 cells were counted foreach sample. When siRNA applied, siRNA oligos were cotrans-fected with pCBASce.

In vivo NHEJ and HDR assaysIn vivo reunion assays were based on the reactivation of line-

arized plasmid as previously reported (16). Briefly, cells were

Feng et al.

Mol Cancer Ther; 15(5) May 2016 Molecular Cancer TherapeuticsOF2




pretreatedwith SAHA for 72hours orwith cisplatin for 24hours. Atotal of 1 � 105 cells were then collected and cotransfected with1.2 mg I-SceI–linearized EJ5-GFP or DR-GFP substrates and 0.5 mgcircular pDsReD-Express2-N1 (as transfection control) by usingelectroporation. After replating, cells were treated with 2 Gy of IR.Flow cytometry analysis was performed 72 hours later withFortessa Flow Cytometer (Fluofarma). The ratio of GFP-positivecells to DsRed-positive cells was used as a measure of relativeactivity for NHEJ or HDR.

Flow cytometry analysisCells were detachedwith stempro accutase (Life Technologies),

and washed twice with PBS. Cells were then stained with phyco-erythrin (PE)-conjugated anti-Sox 2, anti-Oct3/4, anti-Nanog, oranti-CD133 antibodies, or costained with PE-conjugated anti-CD24 and FITC-conjugated anti-CD44 antibodies (BD Bios-ciences). In the process for staining of Sox 2, Oct3/4, and Nanog,BDPerm/Washbuffer was also used as permanufacturer's instruc-tion. PE- or FITC-positive cells were quantified by flow cytometry.

For cell-cycle analysis, cells were collected and fixed with 80%ethanol, stained with propidium iodide, and analyzed by flowcytometry. FlowJo software (FlowJo data analysis software) wasused for cell-cycle analysis.

Immunoblot assayCell lysates were prepared in RIPA buffer with mild sonication

and subjected to SDS-PAGE gel for immunoblot assays. For assaysof acetyl-histone H4, 1mmol/L TSA, and 5mmol/L nicotinamidewere added into RIPA buffer.

Invasion assayA total of 5 � 104 cells in growth medium containing 1% FBS

were seeded in 1� BME (Trevigen) coated 8.0-mm pore size cellculture inserts (for 24-well plate; Millipore). Complete growthmedium containing 10% FBS was placed outside the chambers,and cellswere allowed to invade toward the attractant of full-serummedium.Chamber filter processing and visualization/quantitationof invasion was performed as described previously (19).

Primary tumor tissue cultureFour primary tumor tissues were collected from surgical speci-

mens of lung cancer patients at Affiliated Hangzhou Hospital ofNanjing Medical University. Informed consents were obtainedfrom all patients before operation and the procedure wasapproved by the Medical Ethics Committee. Tumor tissues wereimmediately cut into 4mmx4mmsections, and randomly platedinto 6-well dishes precoated with 2 mL methylcellulose medium(STEMCELL Technologies). Complete medium containing SAHAand cisplatin were then added. Twenty-four hours later aftercisplatin treatment, tissue sections were irradiated or left untreat-ed. Immunofluorescence analysis for g-H2A.X staining was per-formed in frozen tissue sections.

Tumor growth analysis and tumor-initiating testAnimal protocol (Project # SCXK2008–0016) for tumor-initi-

ating test and tumor growth assay was reviewed and approved bythe Institute Animal Ethical Committee at Zhejiang ExperimentalAnimal Center and Zhejiang Academy ofMedical Sciences (Hang-zhou, China).

For tumor growth assay, 1 � 106 of H460 cells in 0.2 mLof HBSS/Matrigel (Life Technologies) mixture (1:1 V/V) were

inoculated subcutaneously into the right thigh of 4- to 6-week-old female athymic nude mice (Charles River Laboratories).When tumor volumes reached a size of 20 to 50 mm3, mice wererandomly grouped into 6 groups (n¼ 6–9) and receive followingtreatments: (i)DMSO for 5days as control; (ii) IR (5Gy) x1onday3 after initiating the treatments; (iii) SAHA (20 mg/kg/d) for 5days; IR was delivered on day three after SAHA injection; (iv)cisplatin (1 mg/kg/d) for 2 days at day 2 and day 3; IR wasdelivered on day three after cisplatin injection; (v) combinationtreatment of SAHA and cisplatin; (vi) combination treatment ofSAHA, cisplatin, and IR. All chemicals were delivered with intra-peritoneal injection, and mice were irradiated locally on the rightthigh using a collimator with a 30-mm opening. Tumors weremeasured biweekly and tumor volumes were determined fromcaliper measurements of tumor length (L) and width (W) accord-ing to the formula (L � W2)/2.

Tumor-initiating test was conducted following the described inMaterials and Methods (20). Briefly, H460 cells were treated with1 mmol/L SAHA or DMSO for 72 hours. Freshly prepared cellswere resuspended in serum-free RPMI1640/Matrigel mixture (1:1v/v), and 500 cells were inoculated subcutaneously to bilateralfranks of same female NOD/SCID mouse (6 weeks old; CharlesRiver Laboratories). The mice were euthanized 4 weeks aftertumor cell injection and tumors were excised. Tumor sizes weremeasured using a caliper and hematoxylin and eosin (H&E)staining was performed for validation of formed tumors.

Statistical analysesStatistical analyses were performed using the Student t test. A P

value of <0.05 was considered as significant (�).

ResultsCisplatin-induced and SAHA-induced radiosensitizationsinvolve distinct compromised DNA damage repair pathways

It has been previously reported that cisplatin treatment doesnot affect NHEJ activity, however, complex cisplatin-DSB lesionsdirectly impair cellular NHEJ (11, 21). On the other hand, ourstudies showed that exposure to SAHA reduced expression ofRad51, a key element involved in HDR pathway, and decreasedHDRactivity in irradiated cancer cells (14, 16). These observationssuggest that cisplatin and SAHAmay act on distinct DNA damagerepair pathways, both leading to the potential of enhancingcellular sensitivity to IR treatment.

To test this, we first evaluated the effects of cisplatin and SAHAon different DNA damage repair machineries. In a pilot study, wefound that SAHAat aminimal concentration of 500nmol/L couldinduce acetylation of histoneH4 inU2OS-DR cells, and treatmentwith 2 mmol/L of cisplatin for 24 hours induced G2–M accumu-lation in U2OS-EJ cells. However, we observed that exposure toeither 2 mmol/L SAHA for 72 hours or 5 mmol/L cisplatin for 24hours caused obvious cytotoxicity in nontumorigenic humanmammary epithelial MCF-10A cells, as shown for decrease ofclonogenic survival (Supplementary Fig. S1). We thus used 1mmol/L of SAHA for 72 hours and 2 mmol/L of cisplatin for 24hours as treatment schedules. With DNA damage repair assaysusing U2OS reporter cells, we detected significant reduction ofHDR activity in cells with SAHA treatment (Fig. 1A). We alsoobserved that exposure to SAHA dramatically decreased Rad51protein level, and knocking down of Rad51 by siRNA transfectioneliminated the inhibitory effect of SAHA on HDR activity

SAHA Enhances Radiosensitization of Cisplatin in Lung Cancer

www.aacrjournals.org Mol Cancer Ther; 15(5) May 2016 OF3




(Fig. 1B). However, no obvious changes were detected in anyreporter cell lines when cells were treated with cisplatin. Inaddition, we noticed that cisplatin treatment did not change thereunion frequency of transfected linear EJ5-GFP plasmid DNA,which indicates the in vivoNHEJ repair capability and is eventuallyaffected by silencing of DNA-pKcs expression (Fig. 1C).

We next tested the radiosensitization effects of cisplatin andSAHA in NHEJ- and HDR-deficient CHO and hamster lungfibroblast cells. We performed median effect analyses in thesecell lines with treating cells with constant ratios of correspondingLD50 values of IR, SAHA, and cisplatin that were determinedwith colony formation assay (Fig. 1D and Supplementary Fig. S1).Our results showed that treatments with cisplatin and SAHAboth induced synergistic radiosensitization (as indicated ofCI values < 1) in control V79 and AA8 cells, and thesesynergistic radiosensitization effects were further enhanced cor-respondingly in BRCA2-defective VC8 for cisplatin (Fig. 1E) and

in DNA-pKcs–defective V3 cells for SAHA (Fig. 1F), with obvi-ously reduced CI values. However, CI values stayed approachingto near 1 in cisplatin-treated V3 cells and in SAHA-treated VC8cells, indicating that the synergisms for radiosensitization effectswere eliminated in NHEJ-deficient cells for cisplatin and in HDR-deficient cells for SAHA.

These results validated that cisplatin can induce synergisticradiosensitization through impaired NHEJ (without affectingNHEJ activity), and SAHA compromises HDR, which leads toincreased sensitivity of cells to IR treatment.

Effects of SAHA and cisplatin on DNA damage repair pathwaysin human lung cancer cells in response to IR treatment

In human lung cancer H460 and H1299 cells, treatment with1mmol/Lof SAHA for 72hours orwith2mmol/Lof cisplatin for 24hours also showed their biologic effects on histone acetylation

0

0

0

AA8

V3

VC3 VC8V79

Sur

viva

l fra

ctio

n (L

og)

V79

VC8V79

AA8

V3

AA8

V3

AA8

V3

AA8

V3

0

0.4

0.8

1.2

0

0.01 0.1 1 10 0.01 0.1 1 10

0.4

0.8

1.2

0 00 0.2

Fraction of cells affected (cytotoxicity) Fraction of cells affected (cytotoxicity)

0.4 0.6 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0

0.5

1.0

1.5

Com

bina

tion

inde

x

Com

bina

tion

inde

x

0

V79

+Cisplatin

+Cisplatin +SAHA

+SAHA

VC8

V79

VC8

0.5

1.0

1.5

0

0.5

1.0

1.5

0

0.5

1.0

1.5

0.4

0.8

1.2

0

0.4

0.8

1.2IR (Gy):

100

10–1

10–2

10–3

100

10–1

10–2

10–3

10–4

2 4 6 8

1

%G

FP

+ C

ells

%G

FP

+ C

ells

%G

FP

+ C

ells

2

3

4

0 24 48 72 Hours

Red51

β-Actin

VehicleSAHA

siRNA-Control

siRNA-Control siRNA-DNA-pKcs

siRNA-Rad51

DMSO

SAHA (1 μmol/L):

SAHA Cisplatin

Cisplatin (μmol/L) SAHA (μmol/L)

DMSO

DMSO

SAHA Cisplatin

Cisplatin

NHEJ

Alt-NHEJ

siR

NA

/co

ntro

l

siR

NA

/D

NA

-pK

cs

siR

NA

-R

ad51

Rad51

β-ActinP = 0.0019

siR

NA

- co

ntro

l

DNA-pKcs

P = 0.026

β-Actin

SSA

HDR0.56%

0.56% 0.38% 0.03%

1.83%

0.04%

0.04%

1.68%

0.56% 0.09%

2.72%0

0

0.2

0.4

0.6

DMSO

siRNA/Control

siRNA/DNA-pKcs

Cisplatin

P = 0.307

0.1

0.2

0.3

0.4

Rel

ativ

e E

Jac

tivity

(10

0%)

Rel

ativ

e E

Jac

tivity

(10

0%)

2.38%

P = 0.0232

0

0

1

1

2

2

3

3

4

45

2468

10

0

0.4

0.8

1.2A

B

DE F

C

Figure 1.Effects of SAHA and cisplatin on DNA damage repair. A, U2OS reporter cells were used to determine the effects of SAHA (1 mmol/L) or cisplatin (2 mmol/L)on distinct DNA DSB repair pathways. Repair of I-SceI–induced DSB was quantified by flow cytometry as described in Materials and Methods. B, inhibitoryeffect of SAHA on HDR is dependent on Rad51 protein. Top, Western blot analysis showing the effect of SAHA on Rad51 protein expression in U2OS-DR reporter cells. b-Actin was included to verify the protein equal loading; bottom, U2OS-DR reporter cells were transfected with siRNA-Rad51, and the effect ofknocking-down Rad51 expression on SAHA-inhibited HDR was measured with flow cytometry. C, in vivo reunion assay for relative NHEJ activity. U2OScells were transfected with or without siRNA-DNA-pKcs, and were then treated with 2 mmol/L cisplatin for 24 hours followed by cotransfection of linearized(with I-SceI) EJ5-GFP plasmid and control pDsRed. Twenty-four hours later, cells were irradiatedwith 2Gy IR, and recycled EJ5-GFPwas counted by flow-cytometryanalysis 48 hours post-IR (left). Graphs show the changes of relative NHEJ activity. D, clonogenic survival analysis was performed to determine LD50 valuesfor IR, cisplatin, and SAHA in paired cell lines. E and F, median effect analyses. Combination indexes (CI) were determined for combined treatment of IRwith cisplatin(E) or with SAHA (F). Data represent the average of three independent experiments.

Feng et al.





(for SAHA) or G2–M accumulation (for cisplatin, SupplementaryFig. S2). As expected, treatment with cisplatin alone had noobvious effects on expressions of NHEJ-relative DNA-pKcs, Ku70,and Ku80 proteins, or HDR-relative Rad51 protein in both H460and H1299 cells, and did not cause any change in either HDR orNHEJ activity in these cells receiving IR treatment, as determinedwith in vivo reunion assay. Treatment with SAHA or SAHA com-bining with cisplatin, however, obviously decreased Rad51 pro-tein expression. Exposure to SAHA also decreased HDR activity inboth irradiatedH460 andH1299 cells (Fig. 2A–C; SupplementaryFig. S2).

Interestingly, we noticed that SAHA treatment decreased Ku80protein expression and reduced NHEJ activity in H460 cells.However, no such changes were observed in H1299 cells

(Fig. 2A and C; Supplementary Fig. S2). Of note, H460 cellsexpress wild-type p53 and H1299 cells express no p53 protein(Supplementary Fig. S2).Mutation of tumor-suppressor p53 geneis one of the most significant molecular events in lung cancers,occurring in about 50% of NSCLC, and clinical studies suggestthat NSCLC with mutant p53 carries a worse prognosis and maybe relatively more resistant to chemotherapy and radiation (22).To verify whether observed effects of SAHA on Ku80 and NHEJactivity correlated with p53 status, we used H1299 cells withengineered expression of wild-type p53, and found that expres-sion of wild-type p53 protein resulted in decreased Ku80 expres-sion and reduced reunion frequency for linear pimEJ5GFP inirradiated H1299 cells when cells were treated with SAHA(Fig. 2D). In addition, our results from assay using DR reporter

E

DMSOSAHA

01234

%G

FP+

Cel

ls HDR

P = 0.0010 P = 0.0042

5

siR

NA

/C

ontro

l

siR

NA

/p5

3

p53

β-Actinβ-Actin

DNA-pKcs

Ku70

Ku80

Rad51

β-Actin

A

H460 H1299

DM

SO

SA

HA

Cis

plat

in

SA

HA

+C

ispl

atin

DM

SO

SA

HA

Cis

plat

in

SA

HA

+C

ispl

atin

DM

SO

DM

SO

SA

HA

SA

HA

Ku80Rad51

p53

pcD

NA

3.1

/EV

pcD

NA

3.1

/Wt-p

53

Rel

ativ

e E

J ac

tivity

(100

%)

0

0.4

0.8

1.2

P = 0.2612

DMSO SAHA

pcDNA3.1

0

0.4

0.8

1.2

P = 0.0122

DMSO SAHA

pcDNA3.1/Wt-p53

Rel

ativ

e E

J ac

tivity

(100

%)

D pcDNA3.1/Empty vector

pcDNA3.1/Wt-p53

DMSO

SAHA

0.07% 0.21%

0.02% 0.09%

1.93% 3.61%

1.11% 2.89%

1.23% 2.39%

0.57% 1.10%

Cisplatin SAHA+Cisplatin

DMSO SAHA

C

1.0 P = 0.1480.80.60.40.2

0Rel

ativ

e E

J ac

tivity

(100

%)

P = 0.004 P = 0.003

DMSO SAHA Cisplatin SAHA+Cisplatin

H460

0.08%2.49% 2.49% 0.23%

2.88% 4.48%

0.42% 0.35%2.92% 2.08%

4.33%3.85%

DMSO SAHA

SAHA+Cisplatin

Cisplatin

0.6

0.4

0.2

0

P = 0.011P = 0.214

P = 0.001

Rel

ativ

e D

R

activ

ity (1

00%

)


B0.12%1.42% 0.11%0.56%

4.01% 3.24%

1.36%

4.00%

0.08%0.54%

3.20%

0.13%

DMSO SAHA

Cisplatin SAHA+CIsplatin

0.3

0.2

0.1

0

P = 0.001

P = 0.620

P = 0.004

Rel

ativ

e D

R

activ

ity (1

00%

)


0.58% 0.08% 0.06%0.36%

2.56% 2.74%

0.68%

3.03% 3.31%

0.42% 0.09%0.08%

H460 H1299

Figure 2.Effects of SAHAand cisplatin on distinct DNAdamage repair pathways in human lung cancer cells. A,Western blot analysis. Cellswere treatedwith 1mmol/L SAHA for72 hours or with 2 mmol/L of cisplatin for 24 hours, or with combined treatment of SAHA and cisplatin as described in Materials and Methods. B and C,in vivo reunion assays for relative HDR (B) and NHEJ (C) activity. Cells were treated with SAHA, cisplatin, or combined treatment as indicated. Cells were thencotransfected with linearized (with I-SceI) DR-GFP (B) or EJ-GFP (C) plasmid and control pDsRed. 2 Gy of IR was delivered 24 hours later after transfection.Recycled DR-GFPwas counted by flow cytometry analysis 48 hours post-IR (top). Graphs showing the changes of relative HDR (B) andNHEJ (C) activity. D, effect ofp53 expression on SAHA-inhibited NHEJ activity. H1299 cells with or without engineering expression of wild-type p53 protein were treated with 1 mmol/L SAHAfor 72 hours, and Western blot analysis (top right) and in vivo reunion assay (left) were then performed as described above. Graphs (bottom right)showing the changes of relative NHEJ activity. E, effect of knocking down p53 on SAHA-inhibited HDR activity. U2OS HDR reporter cells were cotransfected withpCBASce and siRNA-p53, or siRNA-A as control. Repair of I-SceI–induced DSB was quantified by flow cytometry as described in Materials and Methods.Western blot analysis showing the reduced p53 expression in siRNA-p53 transfected cells. Data represent the average of three independent experiments.






cell showed that siRNA knocking down of p53 did not affectSAHA-inhibited HDR activity (Fig. 2E).These results thus sug-gested a potential that, in addition to its effect on HDR, whichis p53-independent, SAHA may also reduce NHEJ activity inirradiated cancer cells in a p53-dependent manner. It is neededto be indicated that we only observed slight but not significantdecrease in NHEJ activity for SAHA treatment in EJ5-reportedcells as shown in Fig. 1A, and this may be caused by the shorterexposure time of SAHA before activation of the NHEJ pathwayin the assay.

SAHA enhances the radiosensitization effect of cisplatin in lungcancer cells

We and others have shown the potential cross-talks amongDNAdamage repair pathways and compromising bothNHEJ andHDR pathways led to enhanced synergistic radiosensitivity incancer cells (16, 23). To test whether SAHA may coordinate withcisplatin on radiosensitivity, we examined the combination effectof SAHA and cisplatin on clonogenic survivals of H460 andH1299 cells receiving a single clinical radiation dose (2 Gy). Wefound that pretreatment with SAHA or cisplatin significantlydecreased clonogenic survival of irradiated cells: For H460, thesurvival fraction decreased from 74.21� 7.79 to 27.94� 5.55 forSAHA treatment (P¼ 0.0011), and to 35.33� 7.40 (P¼ 0.0033)for cisplatin treatment; for H1299, the survival fraction decreasedfrom 78.40 � 8.87 to 41.59 � 3.14 for SAHA treatment (P ¼0.0025), and to 38.85�2.27 for cisplatin treatment (P¼0.0017),respectively. As expected, our results showed that combinedtreatment of SAHA and cisplatin further led to dramaticallyenhanced depression on clonogenic survival for both H460 andH1299 cells (P all <0.05 when compared with IR treatment or tothe treatment of each single agent, Fig. 3A and B). In H1299 cells,however, we noticed that the enhancement effect of SAHA oncisplatin-inhibited clonogenic survival of irradiated cells waseliminated when Rad51 expression was knocked down (Fig.3C). This result indicates a role of Rad51 expression, or HDR, inSAHA-enhanced synergistic effect of cisplatin on radiosensitivityof cells.

We further performed median effect analysis in these cell lines.In this experiment, LD50 values for cisplatin and IR, and the CIvalues for the combination treatment of cisplatin and IR weredetermined in cells exposed to SAHA, or DMSO as control. Ourresults showed that the pretreatment with 1 mmol/L SAHA for 72hours decreased LD50 values of IR from 2.96� 0.31 Gy to 1.60�0.26 Gy (P ¼ 0.004) for H460 cells, and from 3.19 � 0.19 Gy to2.03 � 0.18 Gy (P ¼ 0.002) for H1299 cells, respectively. SAHApretreatment also slightly reduced LD50 values of cisplatin from1.82 � 0.15 mmol/L to 1.60 � 0.10 mmol/L in H460 cells, andfrom 2.74 � 0.43 mmol/L to 1.73 � 0.10 mmol/L in H1299 cells;however, no statistical significances were detected for thesechanges (P ¼ 0.0964 for H460 and P ¼ 0.0166 for H1299,Fig. 3D). Interestingly, previous study showed that SAHAat higherconcentration (2.5 mmol/L) enhanced cisplatin-inhibited tumorcell growth when cells were treated with cisplatin for 48 hours(24). The contradictory observations may be caused by thedifferent concentrations of SAHA and treatment protocol usedfor cisplatin.

Median effect analysis showed that although cisplatin aloneinduced synergistic radiosensitization with CI values of < 1 forbothH460 andH1299 cell lines, the overall CI valueswere furtherremarkably reduced for combination treatments of cisplatin and

IR in cells that were with pretreatment of SAHA, indicating thatSAHApretreatment enhances synergistic radiosensitization effectsof cisplatin in these cells (Fig. 3D; Table 1).

In A549 cells, we also observed that pretreatment with SAHAreduced CI values of cisplatin for its synergistic radiosensitizationeffect (Supplementary Fig. S3).

Effect of the combination treatment of SAHA and cisplatin onpersistence of nuclear g-H2AX foci formed in cells with IRtreatment.

IR produces DSBs in chromosomal DNA leading to phosphor-ylations of C-terminal tails of variant H2AX in chromatin andform g-H2AX foci at DNA break sites. Persistence of g-H2A.Xnuclear foci has been suggested to be an indicator of lethal DNAdamage with nonrepaired DNA DSBs (25, 26). To test whetherSAHA exposure affect the persistence of g-H2A.X nuclear foci inlung cancer cells treated with the combination of cisplatin and IR,we examined the kinetic changes of g-H2A.X foci in H460 cells.While unirradiated H460 cells were predominantly negative forg-H2AX, a homogeneous pattern of discrete nuclear g-H2AX fociwas observed immediately after IR delivery, reaching almost 90%of positive fractions, and then decreased slowly over time. Pre-treatments with SAHA or cisplatin alone, or with combinedtreatment of SAHA and cisplatin, did not cause obvious basallevel changes of nuclear g-H2A.X foci. However, the percentage ofcells with residual g-H2A.X foci at time point of 60 hours post-IRremained higher in cells with pretreatment of SAHA (30.7� 5.27vs. 11.7� 2.38 for control, P¼ 0.005) or of cisplatin (25.3� 5.18,P ¼ 0.014). Notably, in cells treated with the combination ofSAHA and cisplatin, more than half (60.7 � 4.11) of the cellsshowed persistent nuclear g-H2A.X foci after 60 hours post-IR,with statistical significance when compared with those in cellstreated with IR alone (P ¼ 0.0001), or single agent of SAHA (P ¼0.0015) or cisplatin (P ¼ 0.0008; Fig. 4A), indicating that SAHAcould further reduce the cellular capabilities to repair IR-inducedDNA damage in cells with treatment of cisplatin.

In surgical lung cancer tumor tissues cultured in a three-dimen-sional environment, we found the number of cells with g-H2A.Xnuclear foci formed after IR or combined treatment of IR withcisplatin and SAHA varied remarkably among the cases studied.However, we did observe that pretreatment with SAHA causeddramatically enhanced persistence of g-H2A.X nuclear foci intumor tissues that were treated with IR in combination withcisplatin, as shown in Fig. 4B which illustrates the results fromone case studied.

Effect of the combination treatment of SAHA and cisplatin ontumor growth of H460 xenografts receiving radiotherapy

We further determined the effect of the combination treatmentof SAHA and cisplatin on the tumor growth of H460 xenograftsreceiving radiotherapy. In this study, the doses of SAHA andcisplatin were determined with the results of a pilot tolerancesstudy (data not present). In consistence with the results seen invitro, our in vivo data showed that the combination treatmentdramatically increased tumor growth suppression when com-pared with IR alone (Fig. 4C; Supplementary Fig. S4). Of note,we did not observe any unusual symptoms such as weight loss orsickness in mice that were treated with triple therapy protocol,indicating a well tolerance for in vivo triple therapy when radio-therapy was delivered locally at tumor sites.

Feng et al.





SAHA treatment induces differentiation of stem cell–like lungcancer cells and reduces cancer cell invasiveness

Histone deacetylase inhibitors (HDACi) have been character-ized not only as anticancer drugs, but also as cytodifferentiation-inducing agents (27, 28). Of these HDACis, SAHA has also beenreported to inhibit cancer cell invasion (29). We therefore testedthe potential effect of SAHAon differentiation of CSCs and cancercell invasiveness.

In H460 cells, treatment with 1 mmol/L of SAHA for 72 hourssignificantly decreased the percentages of cells with positivestaining of embryonic stem cell marker Oct3/4 (30), Sox II(31), and Nanog (32), and putative CSCs markers CD133þ

(33) and CD24low/CD44þ (34). Exposure to SAHA also increasedthe expressions of differentiation markers, involucrin and synde-can-3 (Fig. 5A and B and Supplementary Fig. S4; refs. 35, 36).

We further assess whether SAHA treatment could change tumor-igenicity of H460 cells in vivo. In tumor-initiating test, we injectedthe same number of H460 cells that received SAHA treatment, orDMSO as control, subcutaneously to bilateral franks of sameanimal. With as few as 500 cells, both control and SAHA-treatedcells formed tumors at all sites (4/4) inNOD/SCIDmice;However,the average volume for tumors formed with SAHA-treated H460cells were significantly smaller than that of control cells at day 25(111.0 mm3 vs. 12.7 mm3, Fig. 5C). Most importantly, SAHA-treated cells had longer latency for forming tumorswhencomparedwith control cells (14� 4 days vs. 22� 2 days). Thus, these resultssuggested that SAHA could induce differentiation of stem cell–likecancer cells and reduce tumorigenicity of H460 cells.

With Transwell invasion assay, we observed that treatmentwith1 mmol/L SAHA for 72 hours remarkably reduced invasiveness of

No IR

IR

No IR

No

IR

IR

DMSO SAHASAHA+Cisplatin

Cisplatin DMSO SAHASAHA+Cisplatin

Cisplatin

1.2

0.8

0.4

0

1.2

0.8

0.4

0

1.2

0.8

0.4

0

1.2

0.8

0.4

0

1.2

0.8

0.4

0

Sur

viva

l fra

ctio

n (1

00%

)

Sur

viva

l fra

ctio

n (1

00%

)

Sur

viva

l fra

ctio

n (1

00%

)

Com

bina

tion

inde

xC

ombi

natio

n in

dex

Sur

viva

l fra

ctio

n (1

00%

)

DM

SO IR

Cis

plat

in

SA

HA

SA

HA

+C

ispl

atin

DM

SO

DM

SO

Cis

plat

in

Cis

plat

in

SA

HA

SA

HA

SA

HA

+C

ispl

atin

SA

HA

+C

ispl

atin

SA

HA

+C

ispl

atin

Cis

plat

in+

IRSA

HA

+IR +

IR

DM

SO IR

Cis

plat

in

SA

HA

SA

HA

+C

ispl

atin

SA

HA

+C

ispl

atin

Cis

plat

in+

IRSA

HA

+IR +

IR

siR

NA

-

siR

NA

-

Con

trol

Rad

51

Rad51

β-Actin

1.2

1.0

0.8

0.6

0.4

0.2

01.2

1.0

0.8

0.6

0.4

0.2

0

Fraction of cells affected (cytotoxicity)

0 0.2 0.4 0.6 0.8 1.0

H460

H460H460

H1299H1299H1299

W/O SAHA

W/O SAHA

W SAHA

W SAHA

W/O SAHA

W SAHA

W/O SAHA

W SAHA

W/O SAHA

W SAHA

W/O SAHA

W SAHA

Cisplatin (μmol/L) IR (Gy):

IR (2Gy)

0.01 0.1 1 10 0 2 4 6 8 10

100

10−1

10−2

10−3

10−4

10−4

10−3

10−2

10−1

100

siRNA-Control siRNA-Rad51

A B

C D

Figure 3.The effect of combined treatment of SAHA and cisplatin in irradiated human lung cancer cells. A and B, clonogenic survival analysis performed for H460 (A) andH1299 (B) cell line as described in Materials and Methods. Top, demonstrative images for colony formation; bottom, graphs showing the ratios of survivalfraction for colonies formed. C, effect of knocking down Rad51 expression on clonogenic survival of H1299 cells with combined treatment of SAHA andcisplatin. H1299 cells were transfected with siRNA-Rad51, or siRNA-A as control. After 24 hours, cells were treated with indicated treatments, and were thenplated for clonogenic survival analysis. 2 Gy of IR was delivered after plating. Western blot analysis showing the decreased expression of Rad51 protein incells transfected with siRNA-Rad51. D, median effect analysis. Clonogenic survival analysis (left) was performed to determine LD50 values for cisplatinand IR in cells treated with or without SAHA as indicated in Materials and Methods. CI values (right) were determined for combined treatment of cisplatinand IR, at constant ratio of LD50 values, in cells with or without pretreatment of 1 mmol/L SAHA for 72 hours.






H460 cells, as shown with the changes of the percentage forinvasive cells (Fig. 5D and Supplementary Fig. S4). Interestingly,our results showed that cisplatin treatment could also reduce thepercentage of invasive cells, and this effectwas further enhanced incells with pretreatment of SAHA.

In H1299 cells, we noticed barely detectable cell fraction withpositive staining of CD24low/CD44þ, and no obvious changeswere observed for the percentages of CD24low/CD44þ and SOX IIcell populations in cells treated with SAHA. However, exposure toSAHA decreased the percentages of cells with positive staining ofCD133, Oct3/4, or Nanog, and induced protein expressions ofinvolucrin and syndecan-3. We also observed that treatment withSAHA dramatically reduced cancer cell invasiveness (Supplemen-tary Fig. S5).

Taken together, our results present here suggest that SAHA andcisplatin compromise distinct DNA damage repair pathways,which lead to enhanced radiosensitivity in human lung cancercells treated with combination of SAHA and cisplatin; treatmentwith SAHA can also induce differentiation of stem cell–like cancercells and reduce lung cancer cell invasiveness. Of note, thesepotential biologic effects of SAHA are not p53 dependent.

DiscussionDespite the tremendous efforts and progress in lung cancer

research, treatment outcomes for nonlocalized NSCLC remainpoor (37). New treatment strategies are urgently needed toimprove survival for advanced NSCLC patients. In this study, weuncovered a potent synergistic radiosensitization effect of cisplat-in and SAHA (vorinostat), a HDACi approved as single-agenttherapy for refractory cutaneous T-cell lymphoma (38), and forphase I and II clinical trials combining SAHA with chemotherapyand radiotherapy for solid tumors, including lung cancer (39, 40),in human lung cancer cells.

Radiotherapy has been a mainstay therapy for lung cancer fordecades. Although traditional radiation was previously morelimited by technology for normal tissue sparing, novel radiother-apy strategy using conformal planning techniques (such asmodern 3D-CRT) are able to deliver high-dose radiation to thetumor target areas while minimizing dose to surrounding tissues,allowing greater radiotherapy dose for early-stage inoperable

NSCLC patients (41, 42), and thus results in improved controlrates. However, further improvements in the therapeutic index ofradiation through dose escalation or the addition of chemother-apy agents that add to the toxicities of the conditioning regimenremain challenging, and in particular, there is a clear need foragents and strategies that will increase the radiosensitivity ofcancer cells to radiation therapy without significant additionaltoxicity for advanced NSCLC patients.

The molecular basis of radiation response is multifactorial.However, the predominant mechanism by which therapeuticirradiation kills most tumor cells is through clonogenic death.In the process, DSBs are regarded as the specific lesions thatinitiate this lethal response (43, 44), and the repair of DSBs isthen critical in determining radiosensitivity (45, 46). Thus, target-ing DNA damage repair pathways has been a strategy for opti-mizing radiotherapy to improve outcomes for cancer treatment(47). To this setting, the results present here showing that SAHAand cisplatin compromise distinct DNA damage repair pathways,and the combination effect of SAHA and cisplatin on radio-sensitivities of human lung cancer cells and xenograft tumorssuggest a clinic impact for novel therapeutic regimens containingSAHA as radiosensitizer for treatment of NSCLC patients.

Of interest, we showed here that SAHA treatment can inducedifferentiation in the population of stem cell–like cancer cells orCSCs. CSCs represent a subpopulation of tumor cells endowedwith self-renewal and multilineage differentiation capacity, withan innate resistance to cytotoxic agents, which thus becomemajorclinical challenges toward the complete eradication of minimalresidual disease in cancer patients (48). CSCs are also likely toplayessential roles in themetastatic spread of primary tumors becauseof their self-renewal capability and their potential to give rise todifferentiated progenies that can adapt to different target organmicroenvironments (49). Thus, targeting CSCs or inducing celldifferentiation in this cell population may help to challengetumor relapse and cancer metastasis. Indeed, we found thatexposure to SAHA significantly lowers the invasive potential ofhuman lung cancer cells in vitro, indicating a potential of SAHAtreatment for targeting cancer cell invasion and metastasis. Inaddition, the CSC differentiation induced by SAHA may alsocontribute to the observed enhancement of radiosensitization incisplatin-treated cancer cells.

Table 1. Median effect analysis for the effect of SAHA on synergistic radiosensitization of cisplatin in human lung cancer H460 and H1299 cells.

CI valuea at inhibition of DRI valueb at inhibition of

H460 cell lineCombination

Ratio 10% 50% 75% 90% 10% 50% 75% 90%

Without SAHA 2.96:1.82 (Gy:mmol/L) 0.78 0.58 0.53 0.51 0.34 0.85 1.35 2.13 (Gy)0.21 0.53 0.83 1.31 (mmol/L)

With SAHA 1.60:1.60 (Gy:mmol/L) 0.79 0.47 0.37 0.29 0.29 0.38 0.42 0.48 (Gy)0.29 0.27 0.42 0.48 (mmol/L)

P values for paired CI 0.9081 0.0026 0.0198 0.0002

CI valuea at inhibition of DRI valueb at inhibition of

H1299 cell lineCombination

Ratio 10% 50% 75% 90% 10% 50% 75% 90%Without SAHA 3.19:2.04 (Gy:mmol/L) 0.93 0.77 0.70 0.64 0.80 1.23 1.52 1.89 (Gy)

0.51 0.78 0.97 1.21 (mmol/L)With SAHA 2.03:1.73 (Gy:mmol/L) 0.92 0.66 0.56 0.48 0.48 0.67 0.79 0.93

0.41 0.57 0.67 0.80P values for paired CI 0.2984 0.0013 0.0054 0.0011aCI < 1, CI ¼ 1 and CI > 1 indicate synergism, additive effect, and antagonism, respectively.bDRI represents the order of magnitude (fold) of dose reduction that is allowed in combination for a given degree of effect compared with the dose of eachtreatment alone. Upper values are for IR (Gy) and lower values are for cisplatin (mmol/L). P values were determined with Student t test. Data from threeindependent experiments.

Feng et al.





DMSO

SAHA

Cisplatin

SAHA+Cisplatin

DMSO

SAHA

Cisplatin

SAHA+Cisplatin

Post-IR (hours): Post-IR (hours):0 7 7 4836 60 0

120

80

40

0DMSO SAHA SAHA+

CisplatinCisplatin DMSO SAHA SAHA+

CisplatinCisplatin

120

80

40

0

7 36 60 487Hours Post-IR: Hours Post-IR:

γ-H

2A.X

+ P

erce

ntag

e (%

)

γ-H

2A.X

+ P

erce

ntag

e (%

)

5,000

4,000

3,000

2,000

1,000

0

Ave

rage

tum

or v

olum

e (m

m3 )

Control

RT

RT+SAHA

RT+Cisplatin

RT+Cisplatin+SAHA

Cisplatin+SAHA

50 10 15 20 Days post-IR

A

C

B

Figure 4.Effect of SAHA on cisplatin-induced persistence of nuclear g-H2A.X foci in irradiated lung cancer cell and in primary lung tumor tissues, and on tumorgrowth in H460 xenografts with radiotherapy. A, representative images of nuclear g-H2A.X foci formed in irradiated H460 cancer cells. Diagram (bottom)showing the changes of cell fractions with g-H2A.X foci in irradiated H460 cells with indicated treatments. B, representative images of nuclear g-H2A.Xfoci formed in irradiated primary lung cancer tumor tissues; diagram (bottom) showing the changes of cell fractions with g-H2A.X foci in irradiatedprimary lung cancer tumor tissues with indicated treatments. C, effect of combined treatment of SAHA and cisplatin on tumor growth in H460 xenograftsin response to radiotherapy of 5 Gy IR. The growth curves represent the average value in each group of 6 to 9 mice. Error bars, one SE.






Involucrin

β-Actin

DM

SO

SA

HA

Syndecan-3

72 hours

Control

Cisplatin

B

DMSO SAHA

24 hours

Control

Cisplatin

D

Control

Cisplatin

48 hours

CDMSO

SAHA

DMSO SAHA

H.E

Nanog+ Nanog+

CD24low

CD44+CD24low

CD44+

CD133+

DMSO SAHA

SOXII+SOXII+

OCT3/4+OCT3/4+

+

ACD133+

3.18% 1.20%

34.3% 16.4%

98.8% 67.6%

85.1% 59.0%

0.07% 0.12% 0.13% 0.18%

%7.77%5.75 %0.22%3.24

Figure 5.SAHA induces differentiation of stem cell–like cancer cells and inhibits cancer cell invasion of H460 cells treated with cisplatin. H460 cells were treatedwith 1 mmol/L of SAHA or DMSO as control for 72 hours, and collected for flow cytometry, Western blot analysis, and tumor-initiating test, or were replatedfor invasion assay. A, flow cytometry analysis for putative cancer stem cell markers. B, Western blot analyses for expressions of involucrin and syndecan-3in H460 cells with or without SAHA treatment. C, tumorigenesis of H460 cells with or without SAHA treatment. Top, images for collected tumors fromtumor-initiating test; bottom, H&E staining results for formed tumors. D, representative images of Transwell invasion assay; 2 mmol/L of cisplatin wasadded to cells 24 hours before plating for invasion assay.

Feng et al.





However, it needs to be indicated that SAHA may cause epi-genetic regulation for gene expression, as we observed that SAHAtreatment resulted in multiple changes of protein expression inthis study. Indeed, HDACi, including SAHA, has shown promisein clinical trials as epigenetic therapy for human malignancies(50). These facts thus necessitate future studies to rule out/rule inalternative mechanisms of cisplatin þ SAHA synergy in additionto effects on DNA damage repair and on CSC differentiation. Onthe other hand, the detail for the role of p53 in SAHA-regulatedDNA damage NHEJ repair also needs to be determined.

Nevertheless, our data demonstrate the synergistic radiosensi-tization effects of SAHA and cisplatin in NSCLC cells and lungcancer xenografts; SAHA can also induce CSC differentiation andreduce cancer cell invasion, indicating potential therapeutic strat-egies with SAHA-containing chemoradiotherapy may furtherbenefit those patients with nonresectable NSCLC in advancedstages (IIIB and IV). However, future studies are needed toevaluate the clinical applicability for SAHA as a part of thechemoradiotherapy regimen and the overall role of Rad51 forthis potential regimen should be further determined.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

DisclaimerThe funders had no role in study design, data collection and analysis,

decision to publish, or preparation of the article.

Authors' ContributionsConception and design: J. Feng, S. Zhang, K. Wu, X. Chen, S. MaDevelopment of methodology: J. Feng, S. Zhang, K. Wu, B. Wang, H. Huang,X. Chen

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): J. Feng, S. Zhang, K. Wu, B. Wang, H. Jiang, L. Ying,H. Huang, X. ChenAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): J. Feng, S. Zhang, K. Wu, H. Jiang, R. Xu, X. ChenWriting, review, and/or revision of the manuscript: J. Feng, S. Zhang, K. Wu,J.Y.C. Wong, X. ChenAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): J. Feng, S. Zhang, K. Wu, L. Ying, X. Zheng,X. Chen, S. MaStudy supervision: X. Chen, S. Ma

AcknowledgmentsThe authors thank Dr. Jeremy Stark from City of Hope Beckman Research

Institute for providing U2OS reporter cell lines and EJ5-GFP/DR-GFP plasmidDNA, and for his helpful comments on DNA damage repair assays.

Grant SupportThis work was supported by the National Nature and Science Foundation of

China (grant 81272611; to S. Ma), Zhejiang Provincial Foundation of NationalScience (No. LZ13H60001; to S. Ma), Natural Science Foundation of ZhejiangProvince, China (Grant LY13H160028; to J. Feng), Zhejiang Provincial Med-icine and Health Science Foundation, China (grant 2013KYA028; to J. Feng),ZhejiangMedical Science Foundation,China (Grant 2014KYA178; to S. Zhang),Hangzhou Key Disease and Discipline Foundation, China (Project20140733Q15; to S. Zhang), Major Science and Technology Innovation Projectof Hangzhou (20112312A01; to S. Ma), and Medical and Scientific ResearchProjects of Hangzhou, China (Project 20150633B42; to K. Wu).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

ReceivedMay 29, 2015; revised January 16, 2016; accepted January 19, 2016;published OnlineFirst February 2, 2016.

References1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer

statistics. CA Cancer J Clin 2011;61:69–90.2. Yang P, AllenMS, AubryMC,Wampfler JA,Marks RS, Edell ES, et al. Clinical

features of 5,628 primary lung cancer patients: experience at Mayo Clinicfrom 1997 to 2003. Chest 2005;128:452–62.

3. O'Rourke N, Roque IFM, Farre Bernado N, Macbeth F. Concurrent che-moradiotherapy in non–small cell lung cancer. Cochrane Database SystRev 2010:CD002140.

4. Jett JR, Schild SE, Keith RL, Kesler KA. Treatment of non–small cell lungcancer, stage IIIB: ACCP evidence-based clinical practice guidelines (2ndedition). Chest 2007;132:266S–76S.

5. Vokes EE, Senan S, Treat JA, Iscoe NA. PROCLAIM: a phase III study ofpemetrexed, cisplatin, and radiation therapy followed by consolidationpemetrexed versus etoposide, cisplatin, and radiation therapy followed byconsolidation cytotoxic chemotherapy of choice in locally advanced stageIII non–small cell lung cancer of other than predominantly squamous cellhistology. Clin Lung Cancer 2009;10:193–8.

6. Auperin A, Le Pechoux C, Rolland E, Curran WJ, Furuse K, Fournel P, et al.Meta-analysis of concomitant versus sequential radiochemotherapy inlocally advanced non–small cell lung cancer. J Clin Oncol 2010;28:2181–90.

7. Belani CP, Choy H, Bonomi P, Scott C, Travis P, Haluschak J, et al.Combined chemoradiotherapy regimens of paclitaxel and carboplatin forlocally advanced non–small cell lung cancer: a randomized phase II locallyadvanced multi-modality protocol. J Clin Oncol 2005;23:5883–91.

8. Koning CC, Wouterse SJ, Daams JG, Uitterhoeve LL, van den Heuvel MM,Belderbos JS. Toxicity of concurrent radiochemotherapy for locallyadvanced non–small cell lung cancer: a systematic review of the literature.Clin Lung Cancer 2013;14:481–7.

9. Le Chevalier T, Arriagada R, Quoix E, Ruffie P,MartinM,Douillard JY, et al.Radiotherapy alone versus combined chemotherapy and radiotherapy inunresectable non–small cell lung carcinoma. LungCancer 1994;10Suppl 1:S239–44.

10. Blackstock AW, Govindan R. Definitive chemoradiation for the treatmentof locally advanced non–small cell lung cancer. J Clin Oncol 2007;25:4146–52.

11. Sears CR, Turchi JJ. Complex cisplatin-double strand break (DSB) lesionsdirectly impair cellular non-homologous end-joining (NHEJ) indepen-dent of downstream damage response (DDR) pathways. J Biol Chem2012;287:24263–72.

12. Oliver TG, Mercer KL, Sayles LC, Burke JR, Mendus D, Lovejoy KS, et al.Chronic cisplatin treatment promotes enhanced damage repair and tumorprogression in amousemodel of lung cancer. GenesDev 2010;24:837–52.

13. Chen X, Wong JY, Wong P, Radany EH. Low-dose valproic acid enhancesradiosensitivity of prostate cancer through acetylated p53-dependentmodulation of mitochondrial membrane potential and apoptosis. MolCancer Res 2011;9:448–61.

14. Chen X,Wong P, Radany EH, Stark JM, Laulier C,Wong JY. Suberoylanilidehydroxamic acid as a radiosensitizer through modulation of RAD51protein and inhibition of homology-directed repair in multiple myeloma.Mol Cancer Res 2012;10:1052–64.

15. Bennardo N, Cheng A, Huang N, Stark JM. Alternative-NHEJ is a mech-anistically distinct pathway of mammalian chromosome break repair.PLoS Genet 2008;4:e1000110.

16. Chen X, Radany EH, Wong P, Ma S, Wu K, Wang B, et al. Suberoylanilidehydroxamic acid induces hypersensitivity to radiation therapy in acutemyelogenous leukemia cells expressing constitutively active FLT3mutants.PLoS ONE 2013;8:e84515.






17. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: thecombined effects of multiple drugs or enzyme inhibitors. Adv EnzymeRegul 1984;22:27–55.

18. Chou TC.Drug combination studies and their synergy quantification usingthe Chou–Talalay method. Cancer Res 2010;70:440–6.

19. Chan CH, Lee SW, Li CF, Wang J, Yang WL, Wu CY, et al. Deciphering thetranscriptional complex critical for RhoA gene expression and cancermetastasis. Nat Cell Biol 2010;12:457–67.

20. Yu C, Yao Z, Dai J, Zhang H, Escara-Wilke J, Zhang X, et al. ALDH activityindicates increased tumorigenic cells, but not cancer stem cells, in prostatecancer cell lines. In Vivo 2011;25:69–76.

21. Boeckman HJ, Trego KS, Turchi JJ. Cisplatin sensitizes cancer cells toionizing radiation via inhibition of nonhomologous end joining. MolCancer Res 2005;3:277–85.

22. Mogi A, Kuwano H. TP53 mutations in non–small cell lung cancer. JBiomed Biotechnol 2011;2011:583929.

23. Allen C, Halbrook J, Nickoloff JA. Interactive competition between homol-ogous recombination and non-homologous end joining. Mol Cancer Res2003;1:913–20.

24. Jin KL, Park JY,NohEJ,HoeKL, Lee JH, Kim JH, et al. The effect of combinedtreatment with cisplatin and histone deacetylase inhibitors on HeLa cells. JGynecol Oncol 2010;21:262–8.

25. Banath JP, Klokov D, MacPhail SH, Banuelos CA, Olive PL. ResidualgammaH2AX foci as an indication of lethal DNA lesions. BMC Cancer2010;10:4.

26. Taneja N, Davis M, Choy JS, Beckett MA, Singh R, Kron SJ, et al. HistoneH2AX phosphorylation as a predictor of radiosensitivity and target forradiotherapy. J Biol Chem 2004;279:2273–80.

27. MarchionD,Munster P. Development of histone deacetylase inhibitors forcancer treatment. Expert Rev Anticancer Ther 2007;7:583–98.

28. Karantzali E, Schulz H, Hummel O, Hubner N, Hatzopoulos A, KretsovaliA. Histone deacetylase inhibition accelerates the early events of stem celldifferentiation: transcriptomic and epigenetic analysis. GenomeBiol 2008;9:R65.

29. Takada Y, Gillenwater A, Ichikawa H, Aggarwal BB. Suberoylanilide hydro-xamic acid potentiates apoptosis, inhibits invasion, and abolishes osteo-clastogenesis by suppressing nuclear factor-kappaB activation. J Biol Chem2006;281:5612–22.

30. de Jong J, Looijenga LH. Stem cell marker OCT3/4 in tumor biology andgerm cell tumor diagnostics: history and future. Crit Rev Oncog 2006;12:171–203.

31. Carina V, ZitoG, Pizzolanti G, Richiusa P, Criscimanna A, Rodolico V, et al.Multiple pluripotent stem cellmarkers in human anaplastic thyroid cancer:the putative upstream role of SOX2. Thyroid 2013;23:829–37.

32. WangML, Chiou SH,WuCW. Targeting cancer stem cells: emerging role ofNanog transcription factor. OncoTargets Ther 2013;6:1207–20.

33. Bertolini G, Roz L, Perego P, Tortoreto M, Fontanella E, Gatti L, et al.Highly tumorigenic lung cancer CD133þ cells display stem-like featuresand are spared by cisplatin treatment. Proc Natl Acad Sci U S A 2009;106:16281–6.

34. Jaggupilli A, Elkord E. Significance of CD44 and CD24 as cancer stem cellmarkers: an enduring ambiguity. Clin Dev Immunol 2012;2012:708036.

35. Chou SC, Azuma Y, Varia MA, Raleigh JA. Evidence that involucrin, amarker for differentiation, is oxygen regulated in human squamous cellcarcinomas. Br J Cancer 2004;90:728–35.

36. PfanderD, Swoboda B, Kirsch T. Expression of early and late differentiationmarkers (proliferating cell nuclear antigen, syndecan-3, annexin VI, andalkaline phosphatase) by human osteoarthritic chondrocytes. Am J Pathol2001;159:1777–83.

37. Chemotherapy for non–small cell lung cancer. CochraneDatabase Syst Rev2000:CD002139.

38. Duvic M, Talpur R, Ni X, Zhang C, Hazarika P, Kelly C, et al. Phase 2 trial oforal vorinostat (suberoylanilide hydroxamic acid, SAHA) for refractorycutaneous T-cell lymphoma (CTCL). Blood 2007;109:31–9.

39. Reguart N, Rosell R, Cardenal F, Cardona AF, Isla D, Palmero R, et al. PhaseI/II trial of vorinostat (SAHA) and erlotinib for non–small cell lung cancer(NSCLC) patients with epidermal growth factor receptor (EGFR) muta-tions after erlotinib progression. Lung Cancer 2014;84:161–7.

40. Treatment of locally advanced non–small cell lung cancer (NSCLC).Updated: September 3, 2015. Available from: https://wwwclinicaltrials-gov/ct2/show/NCT01059552.

41. Kong FM, Hayman JA, Griffith KA, Kalemkerian GP, Arenberg D, Lyons S,et al. Final toxicity results of a radiation-dose escalation study in patientswith non–small cell lung cancer (NSCLC): predictors for radiation pneu-monitis and fibrosis. Int J Radiat Oncol Biol Phys 2006;65:1075–86.

42. Socinski MA, Marks LB, Garst J, Sibley GS, Blackstock W, Turrisi A, et al.Carboplatin/paclitaxel or carboplatin/vinorelbine followed by acceleratedhyperfractionated conformal radiation therapy: a preliminary report of aphase I dose escalation trial from the Carolina Conformal Therapy Con-sortium. Oncologist 2001;6Suppl 1:20–4.

43. Ward JF. The complexity of DNA damage: relevance to biological con-sequences. Int J Radiat Biol 1994;66:427–32.

44. Kastan MB, Bartek J. Cell-cycle checkpoints and cancer. Nature 2004;432:316–23.

45. Dikomey E, Brammer I. Relationship between cellular radiosensitivity andnon-repaired double-strand breaks studied for different growth states, doserates and plating conditions in a normal human fibroblast line. Int J RadiatBiol 2000;76:773–81.

46. Borgmann K, DedeM,Wrona A, Brammer I, Overgaard J, Dikomey E. ForX-irradiated normal human fibroblasts, only half of cell inactivationresults from chromosomal damage. Int J Radiat Oncol Biol Phys2004;58:445–52.

47. Kelley MR, Logsdon D, Fishel ML. Targeting DNA repair pathways forcancer treatment: what's new?Future Oncol 2014;10:1215–37.

48. Sampieri K, Fodde R. Cancer stem cells and metastasis. Semin Cancer Biol2012;22:187–93.

49. Baccelli I, Trumpp A. The evolving concept of cancer and metastasis stemcells. J Cell Biol 2012;198:281–93.

50. Yoo CB, Jones PA. Epigenetic therapy of cancer: past, present and future.Nat Rev Drug Discov 2006;5:37–50.


Feng et al.




Published OnlineFirst February 2, 2016.Mol Cancer Ther Jianguo Feng, Shirong Zhang, Kan Wu, et al.

Small Cell Lung Cancer−in Non Cisplatin on Radiation Sensitivity and Cancer Cell Invasion Combined Effects of Suberoylanilide Hydroxamic Acid and

Updated version

10.1158/1535-7163.MCT-15-0445doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/02/02/1535-7163.MCT-15-0445.DC1

Access the most recent supplemental material at:

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

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://mct.aacrjournals.org/content/early/2016/04/20/1535-7163.MCT-15-0445To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0445

http://mct.aacrjournals.org/content/suppl/2016/02/02/1535-7163.MCT-15-0445.DC1

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/early/2016/04/20/1535-7163.MCT-15-0445


Documents

Combined Effects of Suberoylanilide Hydroxamic Acid and ......Combined Effects of Suberoylanilide Hydroxamic Acid and Cisplatin on Radiation Sensitivity and Cancer Cell Invasion in