zinc finger nucleases targeting the human papillomavirus...

Cancer Therapy: Preclinical

Zinc Finger Nucleases Targeting the Human PapillomavirusE7 Oncogene Induce E7 Disruption and a TransformedPhenotype in HPV16/18-Positive Cervical Cancer Cells

WenchengDing1, ZhengHu1, Da Zhu1, Xiaohui Jiang1, LanYu1, XiaoliWang1, Changlin Zhang1, LimingWang1,Teng Ji1, Kezhen Li1, Dan He2, Xi Xia3, Dan Liu1, Jianfeng Zhou4, Ding Ma1, and Hui Wang1

AbstractPurpose: Cervical cancer is mainly caused by infections of high-risk human papillomavirus (HR-HPV).

Persistent expression of HR-HPV oncogenes E6 and E7 is implicated in malignant transformation. The aim

was to provide proof-of-concept data to support use of zinc finger nucleases (ZFN) targeting HPV E7 to treat

HPV-related cervical cancer.

Experimental Design: We designed and constructed ZFNs that could specifically recognize and cleave

HPV16/18 E7 DNA. We tested the cleavage efficiency of selected ZFN16-E7-S2 and ZFN18-E7-S2 by using

single-strand annealing (SSA) assay. Cell viability and colony formation assays were used to estimate the

inhibition of cell growth that received treatments of ZFNs. Gene disruption of HPV E7 and downstream

genes were examined byWestern blotting. Cell apoptosis assay was used to test the specificity and efficiency

of induction of HPV type-specific apoptosis. We also introduced xenograft formation assays to estimate the

potential of inhibition of HPV-related disease.

Results:We found ZFN16-E7-S2 and ZFN18-E7-S2 disrupted HPV E7 oncogenes in HPV16/18–positive

cervical cancer cells. Both ZFNs effectively led to inhibition of type-specific cervical cancer cell growth, and

specifically induced apoptosis of corresponding HPV16- and HPV18-positive cervical cancer cell lines.

ZFN16-E7-S2 and ZFN18-E7-S2 also repressed xenograft formation in vivo.

Conclusion: ZFNs targeting HPV16/18 E7 could effectively induce disruption of E7 oncogenes and lead

to type-specific and efficient growth inhibition and apoptosis of HPV-positive cells. ZFNs targeting HPV16/

18 E7 oncogenes could be used as novel therapeutic agents for the treatment of HPV-related cervical cancer.

Clin Cancer Res; 20(24); 6495–503. �2014 AACR.

IntroductionCervical cancer is the third most commonly diagnosed

cancer in women worldwide (1). Persistent infection withhigh-risk human papillomavirus (HR-HPV) is consideredthemain etiologic factor for cervical cancers (2). Themalig-

nant transformation abilities of HR-HPVs, especiallyHPV16 and HPV18, are mainly dependent on sustainedexpression of the viral E6 and E7 oncogenes. E6 proteinbinds and degrades tumor suppressor p53, thus inhibitingp53-dependent growth arrest and apoptosis in aberrantproliferative host cells. E7 interacts with tumor suppressorRB1, and frees transcription factor E2F to trans-activate itscellular targets, thereby promoting host cell-cycle progres-sion (3). Interestingly, previous studies also demonstratedthat E7 alone, but not E6, is sufficient to immortalizehuman keratinocytes in vitro (4) and induce high-gradecervical dysplasia in a transgenic mouse model (5). Togeth-er, these data suggested that viral oncogene E7may serve asan ideal target formolecular gene therapy of cervical cancer.

Zinc finger nuclease (ZFN) is an artificial endonucleasewhich is engineered by fusing DNA-binding zinc fingerproteins to the FokI DNA-cleavage domain (6). Cleavageis induced when two custom-designed ZFNs dimerize uponbinding DNA to form an active FokI nuclease. By cleavingspecific target DNA, ZFNs create double-strand breaks(DSB) that are then repaired by the nonhomologous endjoining pathway or, if codelivered with a donor template,repaired by the homology directed repair pathway. As a

1Department of Obstetrics and Gynecology, Tongji Hospital, Tongji Med-ical College, Huazhong University of Science and Technology, Wuhan,Hubei, China. 2Department of Neurology, Tongji Hospital, Tongji MedicalCollege, Huazhong University of Science and Technology, Wuhan, Hubei,China. 3Center for Reproductive Medicine, Peking University ShenzhenHospital, Shenzhen, Guangdong, China. 4Department of Hematology,Tongji Hospital, Tongji Medical College, Huazhong University of Scienceand Technology, Wuhan, Hubei, China.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

W. Ding, Z. Hu, and D. Zhu contributed equally to this article.

Corresponding Authors: Ding Ma, Molecular Cancer Center, Tongji Hos-pital, Tongji Medical College, Huazhong University of Science and Tech-nology, Wuhan 430030, China. Phone: 86-27-83662681; Fax: 86-27-83662681; E-mail: [email protected]; and Hui Wang, E-mail:[email protected]

doi: 10.1158/1078-0432.CCR-14-0250

�2014 American Association for Cancer Research.

ClinicalCancer

Research

www.aacrjournals.org 6495

on July 4, 2018. © 2014 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst October 21, 2014; DOI: 10.1158/1078-0432.CCR-14-0250

http://clincancerres.aacrjournals.org/

genome editing tool, ZFNs have been widely used in severalmodel organisms, such as insects (7), roundworms (8), rats(9), and pigs (10). Moreover, several lines of evidencesuggest that ZFNs could be potential therapeutic agents forhuman gene therapy. For instance, ZFNs targetHIV receptorCCR5 provides robust and heritable protection againstHIV-1 infection both in vitro and in vivo (11), which hadbeen permitted by FDA for clinical trials. Transfusion backof the CD4 T cell contained ZFN-modified CCR5 wasconsidered to be safe and effective (decreasing the bloodlevel of HIVDNA inmost patients; ref. 12). ZFNs disruptingthe promoter function of human T-cell leukemia virus type1 (HTLV-1) specifically killed HTLV-1–infected cells in amodel of adult T-cell leukemia (13). In addition, it wasreported that effective cleavage of vital DNA targets byHBV-specific ZFNs decreased HBV pregenomic viral RNA levelsby 29% (14).

In this study, we constructed ZFNs that target HPV onco-gene E7.We first introduced ZFN FokI domainswith variousmutations for heterodimerization (15), andwe screened thebestmutations in the single-strand annealing (SSA) reportersystem. We then showed that HPV E7-targeted ZFNs withheterodimeric FokIs could efficiently disrupt HPV DNA,resulting in growth inhibition and apoptosis in HPV16/18-positive cervical cancer cells. These therapeutic effects ofZFNs were further confirmed in a xenograft mouse model.Here, we present ZFN targeting of HPV DNA as a noveltherapeutic strategy for the treatment of HR-HPV–relatedcervical cancer.

Materials and MethodsPlasmids and ZFN assembly

Plasmids encoding different zinc finger modules wereobtained from Addgene (Zinc Finger Consortium Expres-sion Vector Kit v1.0). The potential ZFN target sites of

HPV16 and HPV18 E7 were identified using ZiFiT software(http://bindr.gdcb.iastate.edu/ZiFiT/), and themulti-fingerarrays were assembled by restriction digest-based context-dependent assembly (CoDA) or modular assembly asdescribed previously (16, 17). The FokI endonucleasedomain was digested from the pPGK-FokI express vector.Other heterodimeric variants were mutated from wild-typeFokI. These FokIs were cloned into the pcDNA3.1 plasmid(Invitrogen), and the sequences of the plasmids were con-firmed using DNA sequencing. The sequences of FokI var-iants were shown in Supplemental Note S1.

Cell culture and transfectionCervical cancer cell lines SiHa (HPV16 positive), HeLa

(HPV18 positive), C33A (HPV negative), and CaSki (HPV16positive) and human embryonic kidney cell line HEK293(HPV negative) were purchased from ATCC and passaged inour laboratory. All the cell lines were authenticated at Shang-hai PaternityGenetic TestingCenter in June2012, using shorttandem repeat DNA profiling (ABI 3130xl Genetic Analyzer,Life Technologies). All the cells were maintained in DMEMsupplemented with 10% FBS (Gibco) and 100 U/mL pen-icillin and streptomycin (Invitrogen) at 37�C in a humidifiedincubator with 5%CO2. The S12 cell line is an immortalizedhuman cervical keratinocyte cell line that contains integratedHPV16 genomes, which was a generous gift from Pro. Ken-neth Raj (Health Protection Agency). The acquisition of thecell line was permitted by the original owner Pro. MargaretStanley (18, 19). S12was cultured in a 1:3mixture of DMEMand Ham F-12 medium supplemented with 5% FBS, 8.4ng/mL cholera toxin, 5 mg/mL insulin, 24.3 mg/mL adenine,0.5 mg/mL hydrocortisone, and 10 ng/mL EGF. All the cellswere transfected using X-tremeGENE HP DNA TransfectionReagent (Roche) according to the manufacturer’s instruc-tions. The ratio of the reagent to DNA in each cell line wasoptimized inpreliminary experiments. The experimentswereperformed three times in duplicate.

Single-strand annealing recombination reporter assayThe construction of the SSA luciferase reporter plasmid,

pSSA Rep3-1, has been described previously (20). The pSSARep3-1, GZF3-L3, and GZF1-R3 ZFN plasmids were a kindgift from Prof. David Segal (University of California, Davis,CA). Briefly, 25 ng of pSSA reporter plasmid with insertedZFN target sequences, 100 ng of each ZFN plasmid, and25 ng of pRL-TK-Renilla luciferase (Promega) were cotrans-fected into HEK293 cells at 80% confluence in a 24-wellplate. At 48 hours posttransfection, the cells were harvestedand lysed according to the protocol of the Dual-LuciferaseReporter Assay System (Promega). The firefly and Renillaluciferase activity were determined using a microplateluminometer (BioTek). All the experiments were performedthree times in duplicate on different days.

T7E1 assayThe T7E1 assay was performed as previously described

(21). Briefly, the genomic DNA was extracted from cellsusing the DNeasy Blood & Tissue Kit (QIAGEN) according

Translational RelevanceCervical cancer is mainly caused by infections of high-

risk human papillomavirus (HR-HPV). Persistentexpression of HR-HPV oncogenes E6 and E7 is impli-cated in malignant transformation. In the present study,we designed and constructed zinc finger nucleases (ZFN)that could specifically recognize and cleave HPV16/18E7 DNA. ZFN-mediated disruption of HPV16/18 E7DNA directly decreased the expression of E7, inducedtype-specific apoptosis in HPV16/18-positive cervicalcancer cells, and inhibited cervical cancer cell growth invitro. However, no effects were observed in ZFN-treatedHPV-negative cells. Finally, ZFNs targetingHPV16/18 E7were confirmed to have tumor suppression activity in axenograftmodel of cervical cancer. Thus, this study is thefirst to report that HPV-DNA targeting ZFNs could beused as novel therapeutic agents for the treatment ofHPV-related cervical cancer.

Ding et al.

Clin Cancer Res; 20(24) December 15, 2014 Clinical Cancer Research6496




to the manufacturer’s handbook. The genomic regionencompassing the ZFN target site was amplified, purified,melted, and annealed to form heteroduplex DNA. Theannealed DNA (200 ng) was treated with 10 units of T7endonuclease I (New England BioLabs) for 15 minutes at37�C and then analyzed on a 10% TBE polyacrylamide gel.

Western blot analysisCells were lysed on ice for 30 minutes with a lysis buffer

containing 150 mmol/L NaCl, 50 mmol/L Tris (pH¼ 7.4),1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, andprotease inhibitor cocktail. The antibodies rabbit anti-b-actin (60008-1-Ig, Proteintech Group), mouse anti-HPV18 E7 (sc-1590, Santa Cruz Biotechnology), mouseanti-HPV16 E7 (sc-6981, Santa Cruz Biotechnology), andrabbit anti-RB1 (10048-2-Ig, Proteintech Group) were usedfor Western blotting.

Cell proliferation assayCell viability and cell proliferation were determined by

the Cell Counting Kit-8 (CCK-8, Dojindo) according to themanufacture’s manual. Briefly, 2 � 103 cells per well wereseeded in 96-well plates and cultured overnight. For cellviability, cells were transfected with various concentrationsof ZFNs for 24 and48hours. For cell proliferation, cellsweretransfected with the IC50 values of ZFN16-E7-S2 andZFN18-E7-S2 (1,022 ng/mL and 1,230 ng/mL, respectively)for the times as indicated. Then 10 mL of CCK-8 dye was

added to each well and the cells were incubated at 37�C for3 hours, the absorbance was determined at 450 nm using amicroplate reader.

Apoptosis assayAfter infection, cells were trypsinized and double stained

with FITC-conjugated annexin V (annexin V-FITC) andpropidium iodide (PI) using an Annexin V-FITC Apoptosisdetection kit (KeyGEN BioTECH) according to the manu-facturer’s instructions. Then, cells were analyzed usingFACSCalibur (BDBioscience). A total of 20,000 events wereacquired for each sample, and the data were analyzed usingBD CellQuest software.

Colony forming assayA total of 200 cells were plated in triplicate in complete

growth medium in 12-well plates and incubated for 2weeks. Colonies of more than 50 mm in diameter werecounted using an Omnicon 3600 image analysis system.The colonies were stained with 0.04% crystal violet andphotographed.

Animal experimentsFour-week-old specific pathogen-free Balb/c-nu nude

mice were purchased from BEIJINGHFK BIOSCIENCE andhoused at the Experimental Animal Center, Tongji MedicalCollege, Huazhong University of Science and Technology(HUST, Wuhan, China). Six mice were randomly assigned

Figure 1. ZFNs targeting HPV16/18 E7 induced cleavage of viral DNA in HPV16/18-positive cells. A, cleavage activity and toxicity of ZFNs targeting HPV16 E7and HPV18 E7 (Supplementary Table S1) were determined by SSA assays. Top, firefly RLU represented the cleavage activity of ZFNs; bottom, Renilla RLUrepresented the toxicity of the ZFNs. White columns, signal in pSSA reporter with corresponding target sequence; black columns, signal in pSSA reporterscotransfectedwith ZFNs. B, activity of ZFN16-E7 containingwild-typeFokI (wt) or heterodimeric variants of FokIs (S0, S1, S2, andS3)was determined bySSAassays. neg, pSSA-16E7 reporter only. C, activity of ZFN18-E7 containing wild-type FokI (wt) or heterodimeric variants of FokIs (S0, S1, S2, and S3) wasdetermined by SSA assays. neg, pSSA-18E7 reporter only. pos, GZF3-L3 þ GZF1-R3 þ pSSA Rep3-1. �, P < 0.05; ��, P < 0.01, compared with thecontrol reporter by ANOVA testing. D, T7 endonuclease 1 (T7E1) assay of ZFNs cleavage in SiHa and HeLa cells. Cells were untreated (Mock) or transfectedwith 1 mg/mL of ZFN empty vector (neg) or different concentrations of ZFN16-E7-S2 plasmids or ZFN18-E7-S2 plasmids for 48 hours, the genomic DNA ofcells was extracted, PCR was performed with the primers encompassing the target sites, and then the PCR products were purified, annealed, digested withT7E1, and analyzed by gel electrophoresis. The expected cleaved DNA bands are indicated by blank arrows.

ZFN Induces E7 Disruption in HPV-Positive Cells

www.aacrjournals.org Clin Cancer Res; 20(24) December 15, 2014 6497




to each group andwere injected subcutaneously in the rightflanks with 5 � 106 HeLa or SiHa cells. When xenograftsreached about 100 to 150 mm3, we started to complex the10 mg of ZFNs plasmids with TurboFect in vivo TransfectionReagent (#R0541, Fermentas, Thermo Fisher Scientific) andinject intratumorally every 3 days according to the manu-facturers’ manual. The tumor growth was measured every 3days using a digital caliper, and the tumor size was calcu-lated by using the formula: L�W2� 0.5. Tumorweight wasmeasuredwhenmicewere sacrificedonday39 (forHeLa)or45 (for SiHa) after cell implantation. The photographs ofsubcutaneously formed tumors were taken after mice weregiven a lethal overdose of sodium pentobarbital anesthesia.All experimental protocols were approved by the Institu-tional Animal Care and Use Committee of HUST, and thestudy was carried out in strict accordance with the Guide-lines for the Welfare of Animals in Experimental Neoplasia(22).

Statistical analysisThe data were expressed as the mean � SD and analyzed

using Student t test. The experiments were performed threetimes in duplicate. Statistical analysis for Fig. 5 was per-formed using one-way ANOVAwith P¼ 0.05 and assumingequal variances. Statistical analyses were performed withGraphPad Prism 5.

ResultsZFNs targeting HPV16/18 E7 induce cleavage of viralDNA in HPV16/18-positive cells

To design zinc finger motifs that specifically target the E7sequences, we analyzed the E7 sequences of HPV16 and

HPV18 to identify possible binding sites for ZFN pairs, allthese target sequenceswere recheckedbyBLAST (Basic LocalAlignment Search Tool). Five pairs of ZFNs target HPV16 E7and three pairs target HPV18 E7 were designed and con-structed by Modular Assembly or CoDA approach (Supple-mentary Table S1; ref. 17). All of these ZFNs were tested bySSA assay, ZFN-603 (ZFN16 E7, constructed by modularassembly) which targets HPV16 E7 and ZFN-758 (ZFN18E7, constructed by CoDA) which targets HPV18 E7 wereselected for the further study because of the best cleavageactivity without any toxicity (Fig. 1A and SupplementaryTable S1). To enhance the cleavage activity of ZFNs, weintroduced a codon-optimized FokI domain in addition tothe wild-type DNA cleavage domain of the type IIS restric-tion enzyme FokI. Sharkey (S418P, K441E; ref. 23) andothers reported heterodimeric variants of FokI nucleasecontaining variable numbers of mutations, which werenamed S0 (KK: E490K, I538K/EL: Q486E, I499 L; ref. 24),S1 (Sharkey þ KK/Sharkey þ EL), S2 (KKR: KKþH537R/ELD: ELþN496D), and S3 (Sharkeyþ KKR/Sharkeyþ ELD;ref. 15), were also applied. The complete sequences of allFokI variants are provided in the Supplementary Note S1.

Next, to choose the best FokI domain variants and assessthe cleavage activities of the candidate ZFNs, we used amammalian cell-based SSA assay, in which a ZFN targetsequence and a stop codonwere inserted between twodirectrepeat halves of the firefly luciferase gene pSSA Rep3-1(20). A ZFN-induced DSB between the segments allowsefficient SSA homologous recombination using the twohalves as templates, resulting in an active luciferase gene.The Renilla luciferase plasmid was used as an internaltransfection control and a marker for ZFN toxicity.

Figure2. ZFN-mediatedHPV16/18E7genedisruption. A, the representative imagesof gH2AX foci (red) in ZFN16-E7-S2- and si16-E7–treatedSiHa cells and inZFN18-E7-S2- and si18-E7–treated HeLa cells. Cell nuclei were indicated by DAPI staining (blue). Etoposide (0.25 mmol/L) was used as positive control.B, quantification of gH2AX foci in SiHa and HeLa. Three independent cell fields of 30 cells each were counted as average number of gH2AX foci per nucleus.�, P < 0.05. C, the protein levels of HPV E7 and RB1 in ZFN16-E7-S2–treated SiHa and ZFN18-E7-S2–treated HeLa cells. Cells were transfected with1mg/mLZFNsor 50nmol/L of siRNA for 48hours. Theprotein levelsweredetectedbyWesternblotting.b-actinwasusedasa reference control. D, quantitativeanalysis of the protein levels shown in C. �, P < 0.05, compared with parent HeLa cells (n ¼ 3 replications).

Ding et al.





Compared with other FokI variants, ZFN16-E7-S2 andZFN18-E7-S2 showed the best cleavage activity in the pSSAreporter system (Fig. 1B and C). Measurement of thecotransfected Renilla luciferase expression did not revealany significant decrease in signal (Fig. 1B and C), whichmeans low genotoxicity caused by off-target nucleaseactivity. Therefore, ZFN18-E7-S2 and ZFN16-E7-S2 wereselected as candidates for further investigation. Moreover,to explore the effects of ZFN16-E7-S2 and ZFN18-E7-S2 onviral gene disruption, we examined the cleavage activity ofE7 DNA based on the in vitro T7E1 assay. As shownin Fig. 1D, ZFN16-E7-S2 and ZFN18-E7-S2 could inducesignificant cleavage of the E7 genes. These results indicatethat ZFN16-E7-S2 and ZFN18-E7-S2, which target HPV16E7 and HPV18 E7, respectively, have efficient viral DNAcleavage activity.

HPV16/18 E7 gene disruption by ZFN16/18-E7-S2To verify the cleavage effects resulted from genome

editing of HPV E6/E7, we performed immunofluores-cence (IF) staining of gH2AX, which is a sensitive markerfor detecting DSBs. After treatment with ZFN16-E7-S2,the number of gH2AX foci in SiHa were significantlyincreased to 1.53 � 0.10 per nucleus (P < 0.05, comparedwith Vector). Similarly, the number of gH2AX foci inHeLa increased to 20.34 � 0.71 per nucleus after treat-ment with ZFNs (P < 0.05, compared with Vector).

However, both si16-E7 and si18-E7 did not affect theexpression of gH2AX (Fig. 2A and B). It was reported thatSiHa cells contain one to two copies of HPV16 and HeLacells contain 40 to 50 copies of HPV18 (25). The numberof gH2AX foci in ZFNs-treated SiHa and HeLa were nomore than HPV copy numbers, respectively, indicatingthe specificity of ZFNs.

Our previous study showed that RNAi against HPV16E7 led to downregulation of E7 and upregulation ofRB1 (26). To investigate ZFN16/18-E7-S2–mediatedHPV16/18 E7 gene disruption activity, SiHa cells weretransfected with 1 mg/mL ZFN16-E7-S2 or 50 nmol/L ofsiRNA16-E7, and HeLa cells were transfected with 1 mg/mL ZFN18-E7-S2 or 50 nmol/L of siRNA18-E7. HPV E7,RB1 protein expression was detected by Western blotting48 hours after transfection. In this assay, ZFN16/18-E7-S2remarkably decreased HPV16/18 E7 expression inZFN16/18-E7-S2–treated cells. E7 repression resulted inincreased expression of RB1. Similar results were obtainedin siRNA16/18-E7–treated cells (Fig. 2C and D). Theseresults are consistent with those of the previous report(26).

Altogether, these results suggest that ZFN16/18-E7-S2could effectively disrupt the HPV E7 oncogene in HPV16/18-positive cervical cancer cells, leading to repression of theendogenous E7 oncogene and restoration of the RB1 tumorsuppressor pathway.

Figure 3. Knockdownof HPV16/18E7byZFNs inhibited SiHa andHeLa cell growth. The cell viability and cell growthwere determined byCCK-8 assay. For cellviability, SiHa and HeLa were treated with different concentrations of ZFN16-E7-S2 and ZFN18-E7-S2, respectively, for 24 to 48 hours. For cell growth, SiHaand HeLa were treated with ZFNs at the concentration of IC50 value (1,022 ng/mL for ZFN16-E7-S2 and 1,230 ng/mL for ZFN18-E7-S2) for 5 days. Of note,50 nmol/L siRNAs were used as a control. A, dose–response curves of ZFN16-E7-S2–treated SiHa and ZFN18-E7-S2–treated HeLa cells. B, cell growthcurves of ZFN16-E7-S2–treated SiHa and ZFN18-E7-S2–treated HeLa cells. Colony-forming assay was determined by crystal violet staining. C, thecolony-forming assay of SiHa cells after treatment with 2 mg/mL ZFN16-E7-S2 or 50 nmol/L siRNA16-E7 for 24 hours. D, the colony-forming assay of HeLacells after treatment with 2 mg/mL ZFN18-E7-S2 or 50 nmol/L siRNA18-E7 for 24 hours. The columns in C and D represent the average number of colonies ofthree separate experiments; data represent the mean � SD. �, P < 0.05.






Knockdown of HPV E7 inhibits cervical cancer cellgrowth

It has been reported that stable suppression of E6/E7oncogenes expression using specific siRNA induces inhibi-tion of tumor growth both in vitro and in vivo (27, 28). Toevaluate the effect of repression of E7 by ZFNs on tumor cellgrowth, SiHa and HeLa cells were transfected with differentconcentrations of ZFN16-E7-S2 and ZFN18-E7-S2, respec-tively, for 24 to 48 hours. Afterward, the cell viability wastested using a CCK8 assay. Significantly reduced viability ofZFN16/18-E7-S2–treated cells could be observed at con-centrations of more than 500 ng/mL of ZFN16/18-E7-S2(Fig. 3A). The IC50 values of ZFN16-E7-S2 and ZFN18-E7-S2 at 48 hours were calculated to be 1,022 ng/mL and 1,230ng/mL, respectively. As shown in Fig. 3B, both SiHa cellstreated with 1,000 ng/mL of ZFN16-E7-S2 and HeLa cellstreated with 1,200 ng/mL of ZFN18-E7-S2 for 5 daysexhibited significant growth inhibition compared withcontrol groups. Similar results were achieved in siRNA16/18-E7–treated cells.

Moreover, colony-forming assays were used to determineZFN16/18-E7-S2–mediated tumor suppression activity.SiHa and HeLa cells were treated with 1 mg/mL ZFN16-E7-S2 and ZFN18-E7-S2 for 24 hours, respectively. Similarto the results of the siRNA groups, knockdown of E7 byZFN16-E7-S2 and ZFN18-E7-S2 caused a drastic drop in thenumber of clonogenic cells after 2 weeks (Fig. 3C and D).These observations suggest that knockdown of HPV E7 byspecific ZFNs could effectively lead to inhibition of cervicalcancer cell growth.

ZFN16/18-E7-S2 targeting E7 induces apoptosis oftype-specific HPV-positive cells

To study the potential effects of apoptosis induction byZFN16/18-E7-S2 in human HPV-positive cells, SiHa andHeLa cells were treated with ZFN16-E7-S2 and ZFN18-E7-S2, respectively. Cell apoptosiswas quantified48hours laterusing Annexin V and PI staining. As shown in Fig. 4A, SiHacells incubated with different concentrations (500, 1,000,and 1,500 ng/mL) of ZFN16-E7-S2 had an increased apo-ptotic rate. Similar results were obtained with HeLa cellsincubated with different concentrations (600, 1,200 and1,800 ng/mL) of ZFN18-E7-S2.

To further explore the efficacy of ZFN16/18-E7-S2 inthe induction of apoptosis in other cell lines, we cross-transfected SiHa cells with ZFN18-E7-S2 and HeLa cellswith ZFN16-E7-S2. The results showed that neither SiHacells treated with ZFN18-E7-S2 nor HeLa cells treatedwith ZFN16-E7-S2 had an increased apoptotic rate (Fig.4B). To further confirm the specificity of ZFN16/18-E7-S2, CaSki (HPV16 and HPV18 positive, HPV16 predom-inant), C33A (HPV-negative), and S12 (an immortalizedcervical epithelial cell line that contains integrated HPV16DNA) cells were treated with ZFN16-E7-S2 or ZFN18-E7-S2. Interestingly, as shown in Fig. 4C, treatment ofZFN16-E7-S2, but not ZFN18-E7-S2, induced substantialapoptosis of CaSki and S12 cells. This treatment had noeffect on C33A cells (Fig. 4D). Finally, we also used thenormal human embryonic kidney cell line HEK293 todetect the toxicity of ZFNs. We found that neither ZFN16-E7 nor ZFN18-E7 has a role in cell apoptosis in this cell

Figure 4. ZFN16/18-E7-S2 targeting HPV E7 induced apoptosis of type-specific HPV-positive cells. Cells were transfected with different expectedconcentrations or 2mg/mLof ZFN16-E7-S2or ZFN18-E7-S2 for 48 hours. Then, the cellswere collected andstainedwithAnnexin V/PI. The apoptotic ratewasanalyzed by flow cytometry. A, the apoptotic rates of SiHa cells treated with different concentrations of ZFN16-E7-S2 and HeLa cells treated with differentconcentrations of ZFN18-E7-S2. B, the apoptotic rates of ZFN16-E7-S2- and ZFN18-E7-S2–treated SiHa andHeLa cells. C, the apoptotic rates of CaSki andS12 cells. D, the apoptotic rates of HPV-negativeC33A and 293T cells. All experimentswere performed three times; data represent themean�SD. �,P <0.05.

Ding et al.





line (Fig. 4D). These results indicate that ZFN16-E7-S2and ZFN18-E7-S2 could only specifically induce apopto-sis of corresponding HPV16- and HPV18-positive cervicalcancer cell lines.

In vivo assay of tumorigenicity in nude miceTo evaluate the potential effects of ZFN16-E7-S2- and

ZFN18-E7-S2–mediated suppression of tumorigenicityin vivo, we used SiHa and HeLa cell xenograft models inBalb/c nude mice. Nude mice were inoculated subcutane-ously with cells to form xenografts, then ZFN16-E7-S2 andZFN18-E7-S2 were administrated intratumorally usingin vivo transfection reagent. Mice bearing ZFN16-E7-S2–treated SiHa and ZFN18-E7-S2–treated HeLa showedslower growth of xenograft formation (Fig. 5A) and smallertumor size compared with untreated groups (P < 0.05; Fig.5B). Then hematoxylin and eosin (H&E) staining and IHCstainings of PCNA, caspase-3, and CD31were performed toinvestigate the proliferation, apoptosis, and angiogenesis inthe xenograft tumor sections. Both H&E and IHC stainingsshowed that E7-targeted ZFNs resulted in necrosis areas ofvarious sizes in the central of xenograft tissues (Fig. 5C and

D). In addition, compared with control groups, we founddownregulation of proliferation marker PCNA and upre-gulation of caspase-3 in the margins of necrosis areas of theZFNs-treated SiHa and HeLa groups, but not in the non-necrosis areas of these groups (Supplementary Fig. S1). Wedid not observe any effects of ZFN on angiogenesis markerCD31 in the xenograft tumor sections (Fig. 5C and D andSupplementary Fig. S1). Taken together, our results furtherconfirmed the therapeutic effects of ZFN in vivo, but alsosuggested that the efficacy of E7-targeted ZFN could beimproved by in vivo delivery methods, for example, viralpackaging system.

DiscussionHR-HPV infections are the major cause of cervical can-

cers. The malignant transformation abilities of HR-HPVlargely stem from consistent expression of viral oncogenesE6 and E7, making them ideal targets for cancer genetherapy. In this study, ZFNs targeting HPV16/18 E7 DNAdecreased the expression of E7 and increased expression ofRB1 and p53. E7 repression resulted in apoptosis and

Figure 5. In vivo assay of tumorigenicity in nudemice. Balb/c-numicewere injected subcutaneously in the right flankswith 5� 106 SiHa orHeLa cells. Then theZFNs plasmids complexed with in vivo transfection reagents were injected intratumorally when the xenografts reached about 100 to 150 mm3. A, thexenografts weremeasured and recorded every 3 days after treatment with ZFNs. B, the photographs and the estimated tumor size of subcutaneously formedtumors in the untreated and ZFN16/18-E7-S2–treated SiHa and HeLa groups. �, P < 0.01, paired Wilcoxon rank test. C, H&E staining and IHC stainings ofPCNA, caspase-3, and CD31 in untreated and ZFN16-E7-S2- and ZFN18-E7-S2–treated SiHa and HeLa xenografts. The black arrows indicate the necrosisareas. Scale bars, 80mm.D, the average necrosis area and IOD (integratedoptical density) value of PCNA, caspase-3, andCD31 in themargin of necrosisweremeasured using Image-Pro PLUS (Version 6.0). Data represent the mean � SD; �, P < 0.05.






growth inhibition in HPV16/18-positive cervical cancercells.

Several research groups have demonstrated that siRNAtargeting E6/E7 can effectively silence viral oncogenemRNAand induce cell apoptosis (27, 29–31). Although the siRNAapproach has several advantages over traditional drugs intreating multiple cancers, there are several challenges thatmust be overcome (32). In the present study, we attemptedto disruptHPV E7DNAusing ZFNs instead of targeting viralmRNA. First, to engineer maximally active ZFNs, we com-pared the efficiencies of FokI variants. We found that theZFNs containing S2 (KKR: ELD) variants could achievemore effective cleavage than any othermutations, includingS3 (Sharkey þ KKR: Sharkey þ ELD) mutations. Theseresults are not consistent with a previous report that showedthat a combination of Sharkey and KKR: ELD mutationsresulted in an additive effect of ZFN activity (15). It ispossible that the Sharkey mutation not only enhances theFokI cleavage efficiency, but also increases nonspecific tox-icity to target cells. Further experiments are currently ongo-ing to investigate whether extending the recognitionsequence of ZFNs (increasing the number of zinc fingers)could reduce the toxicity of Sharkey mutations. ZFNs withS2 FokI variants with low genotoxicity were selected forfurther experiments.

We then introduced ZFNs into HPV-positive cells andfound that ZFN16/18-E7-S2 could induce significantcleavage of E7 DNA. The disruption to viral DNA directlyled to downregulation of E7 expression at the proteinlevels and restoration of the tumor suppressor genes RB1,resulting in apoptosis and growth inhibition of ZFN-treated HPV16/18-positive cervical cancer cells. We alsocompared the effects of ZFNs and siRNA targeting HPVE7. We found that the therapeutic effects of ZFNs on HPV-infected cells were comparable with those of siRNA.Although several therapeutic siRNAs are already beingevaluated in clinical trials, concerns about their efficacyand safety have been raised (33). For instance, the effectsof siRNAs are transient, and human cells can developresistance to long-term application of siRNA. However,ZFNs may solve this problem by targeting the DNA suchthat disruption of the oncogene E7 by ZFNs is permanentand can be passed to the next generation if the virusreplicates itself.

Specificity is one of the most important concerns for thedevelopment of new drugs. Our data showed that ZFN16-E7-S2, but not ZFN18-E7-S2, induced apoptosis of HPV16-positive SiHa, CaSki, and human cervical keratinocyte S12cells. Moreover, HPV-negative cervical cancer C33A cells

and normal HEK293 cells were insensitive to both ZFN16-E7-S2 and ZFN18-E7-S2. These findings confirmed that theeffects of ZFNs are highly specific to HPV subtype. In thenear future, it may be possible that patients could first betested for HPV infection subtype and then administered thecorresponding subtype-specific ZFNs. Although the resultsshowed that ZFN16-E7-S2 and ZFN18-E7-S2 mediatedsuppression of tumorigenicity in a xenograft model ofcervical cancer, further in vivo studies on ZFN deliverymethods are warranted.

In conclusion, this study shows that ZFNs targeting HPVE7 could efficiently downregulate the expression of the E7oncogene, resulting in apoptosis and growth inhibition incorresponding HPV-positive malignant cells. ZFNs target-ing HPV DNA can be a novel effective strategy for genetherapy of cervical cancer.

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

Authors' ContributionsConception and design: W. Ding, Z. Hu, D. Zhu, J. ZhouAcquisitionofdata (provided animals, acquired andmanagedpatients,provided facilities, etc.): L. Yu, X. Wang, C. Zhang, L. Wang, T. Ji, K. Li,D. He, D. Liu, X. XiaAnalysis and interpretation of data (e.g., statistical analysis, biosta-tistics, computational analysis): W. Ding, Z. Hu, D. Zhu, X. Jiang, X. Xia,J. ZhouWriting, review, and/or revision of the manuscript: W. Ding, Z. Hu,D. Zhu, X. Xia, H. WangAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): X. Jiang, X. XiaStudy supervision: J. Zhou, M. Ding, H. Wang

AcknowledgmentsThe authors thank Prof. David J. Segal (University of California, Davis,

CA) for the gift of the pSSA Rep3-1, GZF3-L3, and GZF1-R3 ZFN plasmidsand Prof. Kenneth Raj (Health Protection Agency, Didcot, United Kingdom)for providing the S12 cell line, which was permitted by the primary ownerProf. Margaret Stanley (Division of Virology, National Institute for MedicalResearch, London, United Kingdom).

Grant SupportThis work was supported by funds from the National Development

Program (973) for Key Basic Research of China (No. 2015CB553903and 2009CB521800 to D. Ma, No. 2013CB911304 to H. Wang, andNo. 2009CB521806 to J. Zhou) and the National Natural Science Fundingof China (NSFC; 81372805 to H. Wang, 81090414 to J. Zhou, 81402158 toZ. Hu, 81471508 and 81101971 to X. Xia, 81402159 to T. Ji, 81301126 toD. He, and 81302266 to K. Li).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received February 15, 2014; revised September 24, 2014; acceptedOctober 1, 2014; published OnlineFirst October 21, 2014.

References1. Arbyn M, Castellsague X, de Sanjose S, Bruni L, Saraiya M, Bray F,

et al. Worldwide burden of cervical cancer in 2008. Ann Oncol2011;22:2675–86.

2. Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA,Shah KV, et al. Human papillomavirus is a necessary cause of invasivecervical cancer worldwide. J Pathol 1999;189:12–9.

3. Moody CA, Laimins LA. Human papillomavirus oncoproteins:pathways to transformation. Nat Rev Cancer 2010;10:550–60.

4. Halbert CL, Demers GW, Galloway DA. The E7 gene of human pap-illomavirus type 16 is sufficient for immortalization of human epithelialcells. J Virol 1991;65:473–8.


Ding et al.




5. Arbeit JM, Howley PM,HanahanD.Chronic estrogen-induced cervicaland vaginal squamous carcinogenesis in human papillomavirus type16 transgenic mice. Proc Natl Acad Sci U S A 1996;93:2930–5.

6. Kim YG, Cha J, Chandrasegaran S. Hybrid restriction enzymes: zincfinger fusions to Fok I cleavage domain. Proc Natl Acad Sci U S A1996;93:1156–60.

7. Bibikova M, Beumer K, Trautman JK, Carroll D. Enhancing genetargeting with designed zinc finger nucleases. Science 2003;300:764.

8. Morton J, Davis MW, Jorgensen EM, Carroll D. Induction and repair ofzinc-finger nuclease-targeted double-strand breaks in Caenorhabditiselegans somatic cells. Proc Natl Acad Sci U S A 2006;103:16370–5.

9. Geurts AM, Cost GJ, Freyvert Y, Zeitler B, Miller JC, Choi VM, et al.Knockout rats via embryo microinjection of zinc-finger nucleases.Science 2009;325:433.

10. Hauschild J, Petersen B, Santiago Y, Queisser AL, Carnwath JW,Lucas-Hahn A, et al. Efficient generation of a biallelic knockout in pigsusing zinc-finger nucleases. Proc Natl Acad Sci U S A 2011;108:12013–7.

11. Perez EE, Wang J, Miller JC, Jouvenot Y, Kim KA, Liu O, et al.Establishment of HIV-1 resistance in CD4þ T cells by genome editingusing zinc-finger nucleases. Nat Biotechnol 2008;26:808–16.

12. Tebas P, Stein D, Tang WW, Frank I, Wang SQ, Lee G, et al. Geneediting ofCCR5 in autologousCD4Tcells of persons infectedwithHIV.N Engl J Med 2014;370:901–10.

13. Tanaka A, Takeda S, Kariya R, Matsuda K, Urano E, Okada S, et al. Anovel therapeutic molecule against HTLV-1 infection targeting provi-rus. Leukemia 2013;27:1621–7.

14. Cradick TJ, Keck K, Bradshaw S, Jamieson AC, McCaffrey AP. Zinc-finger nucleases as a novel therapeutic strategy for targeting hepatitisB virus DNAs. Mol Ther 2010;18:947–54.

15. Doyon Y, Vo TD, Mendel MC, Greenberg SG, Wang J, Xia DF, et al.Enhancing zinc-finger-nuclease activity with improved obligate het-erodimeric architectures. Nat Methods 2011;8:74–9.

16. Sander JD, Dahlborg EJ, Goodwin MJ, Cade L, Zhang F, Cifuentes D,et al. Selection-free zinc-finger-nuclease engineering by context-dependent assembly (CoDA). Nat Methods 2011;8:67–9.

17. Stanley MA, Browne HM, Appleby M, Minson AC. Properties of a non-tumorigenic human cervical keratinocyte cell line. Int J Cancer1989;43:672–6.

18. Bechtold V, Beard P, Raj K. Human papillomavirus type 16 E2 proteinhas no effect on transcription from episomal viral DNA. J Virol2003;77:2021–8.

19. BhaktaMS,SegalDJ. Thegenerationof zincfinger proteinsbymodularassembly. Methods Mol Biol 2010;649:3–30.

20. Kim H, Um E, Cho SR, Jung C, Kim H, Kim JS. Surrogate reporters forenrichment of cells with nuclease-induced mutations. Nat Methods2011;8:941–3.

21. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G, Chaplin DJ,et al. Guidelines for the welfare and use of animals in cancer research.Br J Cancer 2010;102:1555–77.

22. Wright DA, Thibodeau-Beganny S, Sander JD, Winfrey RJ, Hirsh AS,Eichtinger M, et al. Standardized reagents and protocols for engineer-ing zinc finger nucleases by modular assembly. Nat Protoc2006;1:1637–52.

23. Guo J, Gaj T, Barbas CR. Directed evolution of an enhanced and highlyefficient FokI cleavage domain for zinc finger nucleases. J Mol Biol2010;400:96–107.

24. Miller JC,HolmesMC,WangJ,GuschinDY, LeeYL,Rupniewski I, et al.An improved zinc-finger nuclease architecture for highly specificgenome editing. Nat Biotechnol 2007;25:778–85.

25. Meissner JD. Nucleotide sequences and further characterization ofhuman papillomavirus DNA present in the CaSki, SiHa and HeLacervical carcinoma cell lines. J Gen Virol 1999;80:1725–33.

26. SimaN,WangW,KongD,DengD,XuQ,ZhouJ, et al. RNA interferenceagainst HPV16 E7 oncogene leads to viral E6 and E7 suppression incervical cancer cells and apoptosis via upregulation of Rb and p53.Apoptosis 2008;13:273–81.

27. Jonson AL, Rogers LM, Ramakrishnan S, Downs LJ. Gene silencingwith siRNA targeting E6/E7 as a therapeutic intervention in a mousemodel of cervical cancer. Gynecol Oncol 2008;111:356–64.

28. Gu W, Putral L, Hengst K, Minto K, Saunders NA, Leggatt G, et al.Inhibition of cervical cancer cell growth in vitro and in vivo withlentiviral-vector delivered short hairpin RNA targeting human papillo-mavirus E6 and E7 oncogenes. Cancer Gene Ther 2006;13:1023–32.

29. Wu S, Meng L, Wang S, Wang W, Xi L, Tian X, et al. Reversal of themalignant phenotype of cervical cancer CaSki cells through adeno-associated virus-mediated delivery of HPV16 E7 antisense RNA. ClinCancer Res 2006;12:2032–7.

30. Yamato K, Yamada T, Kizaki M, Ui-Tei K, Natori Y, Fujino M, et al. Newhighly potent and specific E6 and E7 siRNAs for treatment of HPV16positive cervical cancer. Cancer Gene Ther 2008;15:140–53.

31. Rampias T, Sasaki C,Weinberger P, Psyrri A. E6 and e7 gene silencingand transformed phenotype of human papillomavirus 16-positiveoropharyngeal cancer cells. J Natl Cancer Inst 2009;101:412–23.

32. Oh YK, Park TG. siRNA delivery systems for cancer treatment. AdvDrug Deliv Rev 2009;61:850–62.

33. CastanottoD, Rossi JJ. The promises andpitfalls of RNA-interference-based therapeutics. Nature 2009;457:426–33.






2014;20:6495-6503. Published OnlineFirst October 21, 2014.Clin Cancer Res Wencheng Ding, Zheng Hu, Da Zhu, et al. HPV16/18-Positive Cervical Cancer Cells

Disruption and a Transformed Phenotype inE7Oncogene Induce E7Zinc Finger Nucleases Targeting the Human Papillomavirus

Updated version

10.1158/1078-0432.CCR-14-0250doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2014/10/21/1078-0432.CCR-14-0250.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/20/24/6495.full#ref-list-1

This article cites 33 articles, 9 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/20/24/6495To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-14-0250

http://clincancerres.aacrjournals.org/content/suppl/2014/10/21/1078-0432.CCR-14-0250.DC1

http://clincancerres.aacrjournals.org/content/20/24/6495.full#ref-list-1

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/20/24/6495


zinc finger nucleases targeting the human papillomavirus...

Documents