PCA-1/ALKBH3 Contributes to Pancreatic Cancer by Supporting … · pancreatic cancer and therefore can be potential therapeutic target. Using RNA interference method in vitro and

Tumor and Stem Cell Biology

PCA-1/ALKBH3 Contributes to Pancreatic Cancer bySupporting Apoptotic Resistance and Angiogenesis

Ichiro Yamato1, Masayuki Sho1, Keiji Shimada2, Kiyohiko Hotta1, Yuko Ueda3, Satoshi Yasuda1,Naoko Shigi3, Noboru Konishi2, Kazutake Tsujikawa3, and Yoshiyuki Nakajima1

AbstractThe PCA-1/ALKBH3 gene implicated in DNA repair is expressed in several humanmalignancies but its precise

contributions to cancer remain mainly unknown. In this study, we have determined its functions and clinicalimportance in pancreatic cancer. PCA-1/ALKBH3 functions in proliferation, apoptosis and angiogenesis wereevaluated in humanpancreatic cancer cells in vitro and in vivo. Further, PCA-1/ALKBH3 expression in 116 patientswith pancreatic cancer was evaluated by immunohistochemistry. siRNA-mediated silencing of PCA-1/ALKBH3expression induced apoptosis and suppressed cell proliferation. Conversely, overexpression of PCA-1/ALKBH3increased anchorage-independent growth and invasiveness. In addition, PCA-1/ALKBH3 silencing downregu-lated VEGF expression and inhibited angiogenesis in vivo. Furthermore, immunohistochemical analysis showedthat PCA-1/ALKBH3 expression was abundant in pancreatic cancer tissues, where it correlated with advancedtumor status, pathological stage and VEGF intensity. Importantly, patients with low positivity of PCA-1/ALKBH3expression had improved postoperative prognosis compared with those with high positivity. Our results establishPCA-1/ALKBH3 as important gene in pancreatic cancer with potential utility as a therapeutic target in this fataldisease. Cancer Res; 72(18); 4829–39. �2012 AACR.

IntroductionPancreatic cancer is one of the most deadly of all types of

cancer in humans. The 5-year survival rate is one of the lowest,at 5% (1). Although systemic treatment including gemcitabinehas recently been used for advanced pancreatic cancer, theeffect of current chemotherapy is only modest (2, 3). Further-more, at present, surgical resection offers the only chance forlong-term survival for pancreatic cancer (4, 5). However, thedisease is so aggressive that only about 20% of patients areindication for surgery at the time of diagnosis (6). Therefore,identification of novel targets and development of new ther-apeutic approaches are required against pancreatic cancer toimprove patient prognosis.Failure of apoptosis is one of the key features of tumor

development (7). It is known that pancreatic cancer is highlyresistant to apoptosis induced by various stimuli includingchemotherapeutic agents (8, 9). Therefore, to reveal the resis-tance mechanism of the pancreatic cancer cells to apoptosis

may lead to develop novel and effective strategies for thetreatment of pancreatic cancer. Recently it has been reportedthat prostate cancer antigen-1 (PCA-1) was highly expressed ina few human cancers (10, 11) and was associated with apo-ptotic resistance of prostate cancer (12). To our knowledge, nostudies have addressed the role of PCA-1 in pancreatic cancer.

PCA-1 has been found to be identical to ALKBH3, onemember of human AlkB homologs (11). AlkB, an Escherichiacoli protein, catalyzes the oxidative demethylation of 1-methy-ladenine and 3-methylcytosine in DNA and RNA (13, 14). Atleast 9 putative human AlkB homologs (ALKBH1-8, FTO) havebeen identified (15–19). AmonghumanAlkBhomologs, PCA-1/ALKBH3 has been reported that its protein structure andcatalytic mechanisms of repairing DNA and RNA are quitesimilar to E. coli AlkB (20–23). Besides these physiological rolesof PCA-1/ALKBH3 in humans, PCA-1/ALKBH3 has beenrecently reported to be highly expressed in human actualcancer including prostate cancer (10) and non–small cell lungcancer (11). The following study has shown that silencing ofPCA-1/ALKBH3 gene on prostate cancer cell by siRNA trans-fection induced apoptosis and suppression of cancer-cellinvasion (24). Conversely, other previous studies have shownthat overexpression of PCA-1/ALKBH3 makes COS-7 cellsresistant to cell death due to an SN2 alkylation agent, methyl-methane sulfonate, acting primarily at the N7-position ofguanine and the N3-position of adenine (13, 25). Taken togeth-er, it may be possible that PCA-1/ALKBH3 is critically involvedin survival and invasion on cancer cells in certain humanmalignancies.

In this study, we hypothesized that PCA-1/ALKBH3 mayhave some roles in apoptosis resistance mechanism in

Authors' Affiliations: Departments of 1Surgery and 2Pathology, NaraMedical University, Kashihara, Nara; and 3Laboratory of Molecular andCellular Physiology, Graduate School of Pharmaceutical Sciences, OsakaUniversity, Suita, Osaka, Japan

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Masayuki Sho, Department of Surgery, NaraMedical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan.Phone: 81-744-29-8863; Fax: 81-744-24-6866; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-12-0328

�2012 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 4829

on July 11, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 23, 2012; DOI: 10.1158/0008-5472.CAN-12-0328

http://cancerres.aacrjournals.org/

pancreatic cancer and therefore can be potential therapeutictarget. Using RNA interference method in vitro and in vivo, weinvestigated the biological roles of PCA-1/ALKBH3 in pancre-atic cancer. Furthermore, we also tried to clarify its clinicalimportance in human pancreatic cancer.

Materials and MethodsReagents, animal, and cell lines

Anti-human Ki67 monoclonal antibodies (MIB-1) were pur-chased from Zymed Laboratories. Anti-VEGF and anti-CD31(PECAM-1) antibodies were purchased from Santa Cruz Bio-technology, Inc.. Female C.B-17/lcr-scid/scid Jcl (SCID) mice(6-week-old) were obtained from CLEA Japan, Inc. All micewere maintained under specific pathogen-free conditions inthe animal facility at Nara Medical University. All experimentswere conducted under a protocol approved by our institutionalreview board. The human pancreatic cancer cell lines, PANC-1and MIAPaCa-2 were obtained from RIKEN BioResourceCenter and cultured in RPMI 1640 supplemented with 10%heat-inactivated FBS.

Preparation of antiseraAnti-PCA-1/ALKBH3 antisera were prepared as described

previously against a synthetic PCA-1/ALKBH3 peptide (aminoacids 64 to 76) as the antigen (10). After a 0.5 mg aliquot ofpeptides was emulsified and injected into mice, blood wascollected at 2-week intervals. The relative activity of antiseraagainst the synthetic peptide was tested by ELISA.

Extraction of total RNAs and real-time reversetranscriptase PCR analysis

Total RNA was isolated using RNAspin Mini (GE HealthcareUK Ltd.) and the first-strand cDNA was synthesized from 1 mgRNA using a ReverTra Ace qPCR RT Kit (TOYOBO) accordingto the instructions of the manufacturer. For real-time reversetranscriptase PCR analysis, cDNA was amplified in TaqManFast Universal PCR Master Mix (2�; Applied Biosystems) withgene-specific primers and probe on the StepOnePlus Real-Time PCR System (Applied Biosystems), according to themanufacturer's instructions. Thermal cycling conditions were95�C for 20 seconds, followed by 40 cycles of 95�C for 1 secondand 60�C for 20 seconds. Real-time PCR experiments for eachgene were carried out on 3 separate occasions. All primer/probe sets were purchased from Applied Biosystems. Theexpression level of the housekeeping gene, b2-microglobulinwasmeasured as an internal reference with a standard curve todetermine the integrity of template RNA for all of the speci-mens. The ratio of mRNA level of each gene was calculated asfollows: (absolute copy number of each gene)/(absolute copynumber of b2-microglobulin).

Preparation of cell lysates and Western blot analysisWe resolved the cell lysates in SDS-polyacrylamide gels and

transferred them onto polyvinylidene difluoride membranes(Millipore, Ltd.), which were then blocked in 5% skim milk atroom temperature for 1 hour. The membranes were incubatedwith the indicated primary antibody for 1 hour, and then

incubatedwith horseradish peroxidase–conjugated antimouseor antirabbit IgG (AmershamPharmacia Biotech).Wedetectedperoxidase activity on X-ray films using an enhanced chemi-luminescence detection system (12).

siRNA transfection of PCA-1/ALKBH3For our transfection analyses, PANC-1 and MIAPaCa-2 cells

were seeded in 6-well plates and transfected eitherwith controlRNA (Santa Cruz Biotechnology) or with 100 nmol/L of siRNAof PCA-1/ALKBH3. Transfections were carried out using theLipofectamine system (Invitrogen) in accordance with themanufacturer's protocol when cells ware achieved about30% confluent. The PCA-1/ALKBH3 siRNAduplexes, generatedwith 30-dTdT overhangs and prepared by QIAGEN, were cho-sen against the DNA target sequences as follows: (PCA-1/ALKBH3 (1) target sequence: 50-CAGAGAGGATATAACT-TATCA-30, PCA-1/ALKBH3 (2) target sequence: 50-ATCGC-TATCATCTTTAGGCAA-30).

Cell viability assayCells were cultured in medium containing FBS for 60 hours.

After incubation, MTS [3-(4,5-dimethyl-2-yl)-5-(3-carboxy-methoxyphenyl)-2-(4- sulfophenyl)-2H-tetrazolium, inner salt]reagent (Promega) was added and optical absorbance at 490nm was measured using a microplate reader as previouslydescribed (26).

Apoptosis detection assayAfter transfection with siRNA, cells were collected and

stained by propidium iodide (PI) and FITC conjugatedAnnexinV according to the manufacture's protocol (TACS Annexin V-FITC kit, R&D system). Then, cells undergoing apoptosis werequantified by measurement of those bound Annexin V butnegative for PI. All experimentswere carried out at least 3 timesin duplicate.

Generation of cells with stable PCA-1/ALKBH3expression

PCA-1/ALKBH3 cDNA was ligated into pEBMulti-Neo vec-tors (Wako Pure Chemicals) at an EcoRV/BamHI site. ALKBH3expression vectors or empty vectors were transfected intoMIAPaCa-2 cells plated at 5 � 104 cells/well using Lipofecta-mine 2000 (Life Technologies). The culture media were chan-ged tomedia containing 900 mg/mL geneticin to select for cellswith stable PCA-1/ALKBH3 expression, 24 hours aftertransfection.

Anchorage-dependent cell growth assayAn anchorage-dependent cell growth assay was carried out

using WST-1, (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disul-fophenyl)-2H-tetrazolium,monosodiumsalt, (Dojindo) accord-ing to themanufacturer's instructions. Briefly, MIAPaCa-2 cells(200 cells/well) were seeded in a 96-well microplate up to avolume of 100 mL/well. After 1, 3, and 5 days, 10 mL ofWST-1/1-methoxy-5-methylphenazinium methyl sulfate solution wasadded to each well. Absorbance was measured at 2 hours afterthe addition using a microplate reader with a test wavelengthof 450 nm and a reference wavelength of 630 nm.

Yamato et al.

Cancer Res; 72(18) September 15, 2012 Cancer Research4830




Anchorage-independent cell growth assayMIAPaCa-2 cells were suspended in DMEM–10% FCS with

0.3% agar (Difco) and plated at 7,500 cells/well in 6-wellplates that were precoated with 0.4% base agar in the samemedium. After 2 weeks of culturing, cells were fixed andstained with 0.05% crystal violet in 50% methanol, and thenumber of colonies formed was counted. Images were thentaken under a microscope. Groups or clusters of cells ofdiameter approximately 0.1 mm or larger were consideredcolonies.

In vitro invasion assayInvasion assays were carried out using BD Biocoat Matrigel

Invasion Chambers (24-well plates with an 8.0-mm pore size;BD Biosciences) according to the manufacturer's instructions.In brief, 5 � 104 cells in serum-free DMEM were plated in theupper chamber. DMEM supplemented with 20% FCS wasplaced in the lower chamber as a chemoattractant. After 24hours of culturing, the invasive cells were fixed and stainedusing Diff-Quick stain solution (Sysmex). The number ofinvasive cells was counted under a microscope.

Microarray analysisFor microarray analysis, aRNA was amplified on the Code-

LinkTM Bioarray (Applied Microarrays, Inc.), according to themanufacturer's instructions. Microarray scanning was doneusing GenePix4000B (Molecular Devices). The data were ana-lyzed using CodeLink TM Expression Analysis v5.0 software(Applied Microarray, Inc.).

Animal experimental protocolIn one in vivo model, cells (4 � 106 PANC-1) were

subcutaneously inoculated on one side of the ventral surfacein the lower flank region of SCID mice. Treatment wasstarted 10 days after tumor implantation. We locally injectedeither control RNA or 10 mmol/L of the PCA-1/ALKBH3siRNA with AteloGene Local Use (total 0.1 mL; Koken Co.)2 times a week for 2 weeks, according to manufacturesprotocol. At 4 weeks after tumor implantation, mice weresacrificed and tumors were taken for further analysis. Inanother in vivo model, cells (1 � 106 PANC-1) were ortho-topically implanted in the pancreas of SCID mice. One weekafter tumor implantation, we locally injected only once

Figure 1. Inhibition of PCA-1/ALKBH3 expression by PCA-1/ALKBH3 siRNA decreased cell proliferation and induced apoptosis in vitro. A, PANC-1andMIAPaCa-2 cellswere transfectedwith PCA-1/ALKBH3siRNAor control RNA. The relative expressionof PCA-1/ALKBH3wasdetermined byquantitativereal-time PCR. The expression of mRNA level of PCA-1/ALKBH3 was strongly reduced in PANC-1 and MIAPaCa-2 cells when transfected withPCA-1/ALKBH3 siRNA for up to 72 hours. PANC-1: n¼ 6 of each group;MIAPaCa-2: n¼ 3 of each group. B, cells were treated as above. Total protein lysateswere extracted and subjected to immunoblotting analysis using antibodies against PCA-1/ALKBH3 and actin. The protein expression of PCA-1/ALKBH3wasalso effectively suppressed in both cell lines. Lane 1, parental line; lane 2, control RNA; lane 3, PCA-1/ALKBH3siRNA(1); lane 4, PCA-1/ALKBH3siRNA(2).C, cells were treated as above. After 60 hours incubation, cell proliferation was determined by MTS assay. Cell proliferation was significantly suppressed byPCA-1/ALKBH3 gene silencing in these cells (n ¼ 4 of each group). D, percentages of apoptotic cells were shown. Quantitation of apoptotic cells by flowcytometric analysis of PCA-1/ALKBH3 siRNA-treated PANC-1 andMIAPaCa-2 cells labeled with Annexin V and PI. Annexin V-positive and PI-negative cellswere considered to be apoptotic cells. Data are presented as means� SD (n¼ 6). ��, P < 0.0001; �, P < 0.001, significant difference when the value of siRNAtreatment was compared with that of the control.

PCA-1/ALKBH3 in Pancreatic Cancer

www.aacrjournals.org Cancer Res; 72(18) September 15, 2012 4831




either control RNA or 20 mmol/L of the PCA-1/ALKBH3siRNA with AteloGene Local Use (total 0.1 mL; Koken Co.Tokyo Japan). At 2 weeks after tumor inoculation, the micewere killed and tumors were removed for further analysis.The tumor volume was calculated according to the formulaV ¼ A � B2/2 (mm3), where A is the largest diameter (mm)and B is the smallest diameter (mm).

In situ detection of apoptosis in tumor tissue sectionsFormalin-fixed and paraffin-embedded 5-mm-thick sections

of all tumor samples were used to identify apoptotic cells byterminal deoxynucleotidyl transferase–mediated dUTP nickend labeling (TUNEL) staining using tumor TACS In SituApoptosis Detection Kit (R&D Systems, Inc.). The apoptoticindexwas calculated as the number of apoptotic cells in 10 highpower fields (�400).

PatientsWeexamined116patientswithpancreatic cancerwhounder-

went surgery atDepartmentof Surgery,NaraMedicalUniversity,between 1996 and 2008. Themedian age of the patients was 65.5years, with a range of 33 to 82 years. Tissues were obtained fromthe resected specimens and then fixed in 10% phosphate-buff-ered formalin and embedded in paraffin. Tumorswere classifiedaccording to the TNM staging system (27). Follow-up was until

death or October 2011. Written informed consent was obtainedfrom patients to use their surgical specimens and clinicopatho-logical data for research purposes.

ImmunohistochemistryFormalin-fixed, paraffin-embedded tissues were cut into

5-mmsections, deparaffinized, and rehydrated in a graded seriesof ethanol. Antigen retrieval was done by heating tissue sectionsusing a Target Retrieval Solution, pH 9.0 (DAKO). To blockendogenous peroxidase, sections were immersed in 0.3% solu-tionofhydrogenperoxide inabsolutemethanol for 10minutes atroom temperature and washed in fresh PBS for 3 times, each of5minutesduration.Purifiedmouseanti-humanPCA-1/ALKBH3antibodydiluted1:200withAntibodyDiluent (DAKO)wasaddedand incubated overnight at 4�C. Sectionswerewashed inPBS for3 times, each of 5minutes duration, and thenweuse EnVisionþ,Mouse/HRP or Rabbit/HRP (DAKO) according to the instruc-tions of the manufacturer. Sections were counterstained withhematoxylin, dehydrated in ethanol, cleared in xylene, andcoverslipped. Authorized pathologists who had no knowledgeof the patients' clinical status and outcome evaluated immu-nohistochemistry for PCA-1/ALKBH3. Percentage of cells pos-itive for PCA-1/ALKBH3 were expressed per >1,000 cells exam-ined. Each sample was classified into 2 groups according tomedianofpositivity.Tocountmicrovessels inmurinepancreatic

Figure 2. Overexpression of PCA-1/ALKBH3 promotes anchorage-independent growth and invasiveness of pancreatic cancer cell. A, PCA-1/ALKBH3overexpression inMIAPaCa-2 cells.MIAPaCa-2 cells were episomally transfectedwith ALKBH3 expression vectors or empty vectors and then selected usinggeneticin. Cell lysateswere immunoblottedwith anti-PCA-1/ALKBH3 and anti-b-actin antibodies. Higher expression of PCA-1/ALKBH3was detected in cellstransfected with pEBMulti-Neo-ALKBH3 than in those transfected with empty vectors. B, anchorage-dependent cell growth of MIAPaCa-2 cellsoverexpressing ALKBH3. Cell growth was assessed using the WST-1 assay on days 1, 3, and 5 after plating under standard culture conditions. Data arerepresented as mean� SD of triplicate experiments. Anchorage-dependent cell growth did not differ between cells overexpressing PCA-1/ALKBH3 and thecontrol cells. C, anchorage-independent cell growth of MIAPaCa-2 cells overexpressing PCA-1/ALKBH3. The transfected MIAPaCa-2 cells were cultured insoft agar for 2 weeks. A higher colony number was observed for the cells overexpressing ALKBH3 than for the control cells. Data are representative of 3independent experiments. D, cell invasion assay usingMatrigel InvasionChambers. Data are represented asmean�SDof triplicate experiments (P¼0.0119).

Yamato et al.





cancer tissue, 5 randomly selected areas were counted at�200magnification, and the summation was calculated.

Statistical analysisResults were expressed asmean values� standard deviation

(SD) or as medians with ranges, and the Student t test, the chi-square test, Fisher's exact test, or the Mann–Whitney U testwere used for evaluating statistical significance. Kaplan–Meiersurvival calculations and the corresponding log-rank testswere carried out to determine differences in survival rates.A P value <0.05 was considered statistically significant.

ResultsSilencing of PCA-1/ALKBH3 inhibits tumor cellproliferation and induces apoptosis in humanpancreatic cancer cell linesTo assess the roles of PCA-1/ALKBH3 in human pancreatic

cancer, we first examined the effects of PCA-1/ALKBH3 down-regulation in vitro. Human pancreatic cancer cell lines, PANC-1and MIAPaCa-2 were selected, because these cell lines express

PCA-1/ALKBH3 mRNA and protein. Both mRNA and proteinexpressions of PCA-1/ALKBH3 were substantially reduced inPANC-1 and MIAPaCa-2 cells, when transfected with PCA-1/ALKBH3 siRNA for up to 72 hours (Fig. 1A and B). Cell prolif-eration for up to 60 hours was significantly suppressed by PCA-1/ALKBH3 gene silencing in these cells (Fig. 1C). Apoptosisdetection assay showed a significant increase in the percentageof AnnexinV-positive andPI-negative cellswithdownregulationof PCA-1/ALKBH3 (Control RNA, vs. PCA-1/ALKBH3 siRNA,PANC-1; 3.5� 0.3% vs. 29.0� 3.0%, P < 0.001, MIAPaCa-2; 1.1�0.5% vs. 20.3� 2.4%,P< 0.001; Fig. 1D).Data suggested that PCA-1/ALKBH3 silencing induced apoptosis and restrained cellproliferation in PANC-1 and MIAPaCa-2 cells.

Overexpression of PCA-1/ALKBH3 promotes anchorage-independent growth and invasiveness of pancreaticcancer cell

To further investigate the role of PCA-1/ALKBH3 inpancreatic cancer cell, we then examined the effect of itsoverexpression in vitro. MIAPaCa-2 cells were episomally

Figure 3. Inhibition of PCA-1/ALKBH3 expression by PCA-1/ALKBH3 siRNA induced substantial antitumor effect in vivo. A, immunohistochemical stainingshowed a significant decrease in PCA-1/ALKBH3 expression was evident in tumor tissue samples obtained from PCA-1/ALKBH3 siRNA-treated micewhen compared with mice given control RNA. B, Western blotting further confirmed that PCA-1/ALKBH3 expression was strongly decreased in vivo by localinjection of PCA-1/ALKBH3 siRNA vs. control RNA. C, SCID mice were inoculated s.c. with 4 � 106 PANC-1 cells (day -10). Mice were then locally injectedeither control RNA (n ¼ 8) or PCA-1/ALKBH3 siRNA with AteloGene (n ¼ 6) on days 0, 3, 7, and 10. D, tumor weight of PCA-1/ALKBH3 siRNA- or controlRNA-treated mice measured at sacrifice on 28 day after tumor cell injection. Each symbol represents an individual mouse; long horizontal lines indicate themean tumor weight at sacrifice. P < 0.0001; control RNA, n ¼ 8 vs. PCA-1/ALKBH3 siRNA, n ¼ 6). E, PANC-1 cells were implanted orthotopicallyinto the pancreas of SCID mice (day -7). Groups of mice were locally injected either control RNA (n ¼ 5) or PCA-1/ALKBH3 siRNA with AteloGene(n ¼ 8) on day 0. F, the tumor volume was calculated as above on days 0 and 7 (relative tumor volume at day 7 after treatment; control RNA, n ¼ 5;PCA-1/ALKBH3 siRNA, n ¼ 8). G, Tumor weight of PCA-1/ALKBH3 siRNA- or control RNA-treated mice measured at day 7 after treatment (control RNA,n ¼ 5; PCA-1/ALKBH3 siRNA, n ¼ 8). �, P < 0.01; ��, P < 0.001; and ��, P < 0.0001, significant difference when the value of PCA-1/ALKBH3 siRNAtreatment was compared with that of the control.






transfected with ALKBH3 expression vectors or empty vectorsand then selected using geneticin. Cell lysates were immu-noblotted with anti-PCA-1/ALKBH3 and anti-b-actin anti-bodies. Higher expression of PCA-1/ALKBH3 was detected incells transfected with pEBMulti-Neo-ALKBH3 than in thosetransfected with empty vectors (Fig. 2A). Next we investi-gated anchorage-dependent cell growth of MIAPaCa-2 cellsoverexpressing PCA-1/ALKBH3. Cell growth was assessedusing the WST-1 assay on days 1, 3, and 5 after plating understandard culture conditions. Anchorage-dependent cellgrowth did not differ between cells overexpressing PCA-1/ALKBH3 and the control cells (Fig. 2B). In contrast, in thecase of anchorage-independent cell growth, a higher colonynumber was observed for the cells overexpressing PCA-1/ALKBH3 than for the control cells, suggesting that ALKBH3is an important molecule that regulates the anchorage-independent growth of pancreatic cancer cells (Fig. 2C).Furthermore, overexpression of PCA-1/ALKBH3 promotedthe invasiveness of MIAPaCa-2 cells (Fig. 2D). These resultssuggested that PCA-1/ALKBH3 plays a significant oncogenicrole in pancreatic cancer cell.

Effect of silencing PCA-1/ALKBH3 in pancreatic cancerin vivo

To evaluate the role of PCA-1/ALKBH3 in pancreaticcancer cell under physiological conditions, we then exam-ined the effect of PCA-1/ALKBH3 siRNA in pancreaticcancer in vivo. PANC-1 cells were subcutaneously inoculatedon SCID mice and treated with PCA-1/ALKBH3 siRNA in thepresence of atelocollagen. By immunohistochemistry, PCA-1/ALKBH3 expression was successfully downregulatedby in vivo PCA-1/ALKBH3 siRNA transfection (Fig. 3A). Thiswas confirmed by Western blot analysis on pancreaticcancer tissues at 4 weeks after tumor implantation.Knockdown efficacy was of high degree (Fig. 3B). PCA-1/ALKBH3 downregulation induced substantial antitumoreffect in vivo and significantly inhibited tumor growth (Fig.3C). And the weights of PCA-1/ALKBH3 siRNA treatedtumors at sacrifice were significantly reduced comparedwith controls (Fig. 3D).

To confirm the above in vivo data, we next used anorthotopic in vivo model. PANC-1 cells were orthotopicallyimplanted on SCID mice and treated with PCA-1/ALKBH3

Figure 4. Effect of targeting PCA-1/ALKBH3 on pancreatic cancer in vivo. A, the representative micrographs show immunohistochemical staining for cellproliferation (Ki67) and apoptosis (TUNEL) in tumor tissues obtained from subcutaneousmodels. A significant decrease in Ki67 staining was evident in tumortissue samples obtained fromPCA-1/ALKBH3 siRNA-treatedmicewhen comparedwithmice given control RNA. Conversely, a significant increase in TUNELexpression, representing apoptosis, was evident in tumors obtained from mice given PCA-1/ALKBH3 siRNA compared with controls. B, the number ofTUNEL-positive cells. Apoptosis was significantly induced in PCA-1/ALKBH3 siRNA treated tumors compared with controls (control RNA, n ¼ 5;PCA-1/ALKBH3 siRNA, n ¼ 5). C, the representative micrographs show immunohistochemical staining for Ki67 and TUNEL-positive cells (arrows) in tumortissues obtained from orthotopic model. A significant decrease in Ki67 staining was evident in tumor tissue samples obtained from PCA-1/ALKBH3siRNA-treated mice when compared with mice given control RNA. Conversely, a significant increase in TUNEL expression, representing apoptosis, wasevident in tumors obtained from mice given PCA-1/ALKBH3 siRNA compared with controls. D, apoptotic cells determined by TUNEL staining weresignificantly higher in PCA-1/ALKBH3 knockdown compare to control (control RNA, n ¼ 5; PCA-1/ALKBH3 siRNA, n ¼ 8).

Yamato et al.





siRNA in the presence of atelocollagen. As shown in Fig. 3E,PCA-1/ALKBH3 expression was also successfully downre-gulated by in vivo PCA-1/ALKBH3 siRNA transfection inthis model. PCA-1/ALKBH3 downregulation induced sub-stantial antitumor effect, thereby resulting in significantinhibition of tumor growth (Fig. 3F). Furthermore, theweights of PCA-1/ALKBH3 siRNA treated tumors at sacri-fice were significantly reduced compared with controls(Fig. 3G).Proliferation activity of subcutaneous PANC-1 tumors

examined by Ki67 staining was significantly suppressed byPCA-1/ALKBH3 knockdown in vivo (Fig. 4A). Then, tumorcells undergoing apoptosis in the subcutaneous modelwere quantified by measurement of nuclear staining forTUNEL. As a result, apoptosis was frequently induced incells with PCA-1/ALKBH3 silencing (Fig. 4A and B). In theorthotopic model, Ki67 staining was also significantly sup-pressed by PCA-1/ALKBH3 silencing (Fig. 4C). Apoptoticcells were significantly higher in PCA-1/ALKBH3 knock-down compare to control (Fig. 4D). Taken together, PCA-1/ALKBH3 silencing reduced cell proliferation and inducedapoptosis in human pancreatic cancer cells in vitro and invivo.

PCA-1/ALKBH3 silencing reduces VEGF expression inhuman pancreatic cancer

Next we explored the other underlying mechanism in effectsof PCA-1/ALKBH3 silencing in tumors. To this end, we usedmicroarray analysis to compare the gene expression profiles ofPANC-1 cells transfected with PCA-1/ALKBH3 siRNA versuscontrol RNA. We identified 106 genes that were 3 times up- ordownregulated in PCA-1/ALKBH3 siRNA (vs. control RNA) cells(Supplementary Table S1). In these genes, we focused on VEGFthat is amajor stimulator of angiogenesis, because angiogenesisis akeymechanisminmanyhumancancers.ByqPCR, themRNAexpression of VEGF was significantly downregulated in PANC-1and MIAPaCa-2 cells by PCA-1/ALKBH3 siRNA transfection invitro (Fig. 5A). And PCA-1/ALKBH3 silencing also reducedprotein expression of VEGF in vitro (Fig. 5B). In addition, VEGFexpressionwas significantlydecreased inPCA-1/ALKBH3siRNAtreated tumors compared with controls in both subcutaneousand orthotopic models as shown by immunohistochemistry(Fig. 5C). Western blotting further confirmed that VEGF expres-sionwas strongly decreased in vivo subcutaneousmodels by thisinjection of PCA-1/ALKBH3 siRNA versus control RNA (Fig. 5D).PCA-1/ALKBH3 downregulation significantly decreased thenumbers of micro vessels in tumors (Fig. 5E and F). Data

Figure 5. Effect of angiogenesis targeting PCA-1/ALKBH3 on pancreatic cancer. A, PANC-1 and MIAPaCa-2 cells were transfected with PCA-1/ALKBH3siRNA. The relative expression of VEGF was determined by quantitative real-time PCR (PANC-1: PCA-1/ALKBH3 siRNA, n ¼ 6; control RNA, n ¼ 6;MIAPaCa-2: PCA-1/ALKBH3 siRNA, n ¼ 3; control RNA, n ¼ 3). ��, P < 0.001; ��, P < 0.0001. B, cells were treated as above. Total protein lysates wereextracted and subjected to immunoblotting analysis using antibodies against VEGF and actin. Lane 1, parental line; lane 2, control RNA; lane 3, PCA-1/ALKBH3 siRNA(1); lane 4, PCA-1/ALKBH3 siRNA(2). C, the representative immunohistochemical staining for VEGF in tumor tissues obtained from mice.A significant decrease in VEGF was evident in tumor tissue samples obtained from PCA-1/ALKBH3 siRNA-treated mice when compared with mice givencontrol RNA both in subcutaneous and orthotopic models. D, Western blotting indicated that VEGF expression was strongly decreased in vivo by localinjection of PCA-1/ALKBH3siRNA vs. control RNA. E, immunohistochemical staining for CD31 in tumor tissues obtained frommice. F, analysis ofmicrovesselcounts in tumors at 28-day after tumor cells injection between control RNA and PCA-1/ALKBH3 siRNA mice. The numbers of CD31 positive vessels withintumors treated with PCA-1/ALKBH3 siRNA were significantly lower than control mice (control RNA, n ¼ 5; PCA-1/ALKBH3 siRNA, n ¼ 5; �, P < 0.01).






suggested that PCA-1/ALKBH3 silencing not only reduced cellproliferation and induced apoptosis, but also decreased angio-genesis in human pancreatic tumor.

Clinical significance of PCA-1/ALKBH3 expression inhuman pancreatic cancer

We evaluated the PCA-1/ALKBH3 expression in 116 actualhuman pancreatic cancer tissues by immunohistochemistry.Positive staining was seen in all examined specimens ofpancreatic cancer. On the other hand, PCA-1/ALKBH3 expres-sion was not observed in noncancer tissues of the pancreas.Median positivity of PCA-1/ALKBH3 was 62.5%. Then allspecimens were classified into 2 groups according to themedian of positivity of PCA-1/ALKBH3 expression (Fig. 6A).We evaluated the correlation of the PCA-1/ALKBH3 expressionwith various clinicopathological findings. The 2 groups did notdiffer significantly with respect to age or histopathologicalgrading. We found that tumors with a high positivity of PCA-1/ALKBH3 expression significantly correlated with advanced Tstatus (T1, T2, T3 vs. T4,P¼ 0.0079) andpathological stage (P¼0.0144; Table 1). These data suggested that PCA-1/ALKBH3expression might be functionally important in tumor progres-sion in pancreatic cancer. Then, we compared postoperativeprognosis between 2 groups. Importantly, tumors with lowpositivity of PCA-1/ALKBH3 expression had significantly bet-ter postoperative prognosis compared with tumors with high

positivity (Fig. 6B). Finally, we evaluated correlation betweenPCA-1/ALKBH3 and VEGF expression in actual human pan-creatic cancer. We scored the intensity of immunostaining ofVEGF as follows: 0, no staining; 1, weak staining; 2, moderatestaining; and 3, strong staining (Fig. 6C), and then classifiedscore 0 and 1 as low intensity, and score 2 and 3 as highintensity. We found that tumors with a high positivity of PCA-1/ALKBH3 expression significantly correlated with high inten-sity of VEGF expression in human cancer (P < 0.0001; Table 1).These data indicated that PCA-1/ALKBH3 might play impor-tant roles in actual human pancreatic cancers and also be apotential therapeutic target for human pancreatic cancer.

DiscussionHumans are continuously exposed to agents that methylate

DNA and RNA, such as tobacco-specific nitrosamines (28) andcellular S-adenosylmethionine (29). These agents may initiateabnormal methylation on genes. When DNA methylation isdysregulated, the harmful methylation contributes to diseasessuch as cancer (30–33). Therefore, the enzymatic functions ofrepairing these abnormalities are critical for maintaining DNAand RNA integrity. Recently, several demethylating enzymeshave been studied very well. In these proteins, it has beenreported that AlkB and its human homologs play an importantrole in demethylation of DNA (13, 14, 19, 20, 22, 23, 34). AmonghumanAlkB homologs, ALKBH3 is known as a uniquemember

Figure 6. Immunohistochemicalstaining of human pancreaticcancer tissue for PCA-1/ALKBH3and survival rate after surgery.A, representative case of highpositivity (Hi) and low positivity(Low) of PCA-1/ALKBH3. Originalmagnification, �100. B, overallsurvival of 116 patients withpancreatic cancer in relation totumor PCA-1/ALKBH3 status.Tumors with high positivity ofPCA-1/ALKBH3 expression hadsignificantly poorer prognosiscompared with tumors with lowpositivity (P ¼ 0.0255). C,immunohistochemical analysis ofVEGF in human pancreatic cancertissues of no (score 0), mild(score 1), moderate (score 2), andstrong (score 3) intensities.

Yamato et al.





to demethylate RNA besides repairing methylated DNA(13, 35–37). Several previous studies suggest that ALKBH3may play physiological key roles in humans.There are only limited studies that have recently described

roles of AlkB family in cancer biology. For instance it has beenreported that PCA-1/ALKBH3 was specifically expressed andcontributed to cell survival and invasion in human prostatecancer (10, 24) and in human non–small cell lung cancer (11).In gastrointestinal tumor, it has been reported that in rectalcancer tumor tissues overexpressed PCA-1/ALKBH3 com-pared with normal rectal tissues, although the function andthe role of PCA-1/ALKBH3 have not been addressed (38). It wasalso reported that PCA-1/ALKBH3 was one of the candidategene associated with the risk of papillary thyroid cancer (39).Nevertheless, the roles of PCA-1/ALKBH3 in carcinogenesiswere largely unknown. In other ALKBHs, ALKBH8 contributesto progression of human bladder cancer (12). The otherhomolog, ALKBH2, downregulation of it increases cisplatinsensitivity in human lung cancer cell line (40). By sharpcontrast, the antitumor activity of ALKBH2 has been reportedin gastric cancer. In that study, overexpression of ALKBH2

significantly inhibited the proliferation of cancer cells, andinduced G1 arrest of the cell cycle, whereas ALKBH2 knock-down promoted cell growth and cell cycle progression ofcancer cells (41). Thus, at present the roles of AlkB family incancers were diverse, complex and controversial. In addition,there is no report that examined the roles of AlkB family inpancreatic cancer. Therefore, in this study, we were intriguedwith PCA-1/ALKBH3 in pancreatic cancer and investigatedprecise roles of PCA-1/ALKBH3 in pancreatic cancer.

We found several important observations. First, we exam-ined the biological mechanism of PCA-1/ALKBH3 on pancre-atic cancer by using siRNA method in vitro. Cell proliferationwas significantly suppressed by PCA-1/ALKBH3 gene silencingin both PANC-1 and MIAPaCa-2 cells. Furthermore, PCA-1/ALKBH3 siRNA transfection induced apoptosis to these pan-creatic cancer cell lines. These might be consistent withprevious reports on prostate cancer (10, 24). On the otherhand, we showed that overexpression of PCA-1/ALKBH3 inMIAPaCa-2 cells increased anchorage-independent growthand invasive potential. These data showed that PCA-1/ALKBH3 had oncogenic significance in pancreatic cancer cells.Second, there is no study to have examined the mechanism ofPCA-1/ALKBH3 knockdown in physiological in vivo model.Then we investigated the in vivo roles of PCA-1/ALKBH3downregulation in pancreatic cancer. We used subcutaneousand orthotopic xenograft tumor models used PANC-1 cells.PCA-1/ALKBH3 knockdown reduced the size of tumors inthese models. As underlying mechanisms, we revealed thatproliferation activity examined by Ki67 was frequently sup-pressed, and apoptosis evaluated by TUNEL was significantlyinduced in cancer tissues in both models.

Next, for a deeper understanding of functions of PCA-1/ALKBH3 in pancreatic cancer, we focused on angiogenesis.Angiogenesis play a key role in tumor growth and metastasis.Since about 40 years ago, it has become to be known that thegrowth of any tumor depends on angiogenesis. Without angio-genesis, the tumor cannot grow beyond a few millimeters indiameter (42). VEGF is the most prevalent and dominantproangiogenic growth factor in the tumor microenvironment(43–45). Althoughmost pancreatic cancers are hypovascular oravascular, the human pancreatic cancer cells reportedly over-express pro-angiogenic molecules such as VEGF (46). We (47)and others (48) have shown that the intratumoral microvesseldensity is an independent prognostic factor in pancreaticcancer patients. Furthermore, in animal models, the anti-angiogenic therapies have been effective in inhibiting pancre-atic tumor development. Therefore, angiogenesis also plays apivotal role in the growth of pancreatic tumors. Recently,antiangiogenic agents include the anti-VEGF antibody beva-cizumab, has shown promise in the treatment of patients withseveral cancers (49, 50). Therefore, we examined involvementof PCA-1/ALKBH3 in VEGF. PCA-1/ALKBH3 knockdown alsoreduced bothmRNA and protein expression of VEGF in PANC-1 and MIAPaCa-2 cells in vitro. These suggested that PCA-1/ALKBH3 correlated with not only apoptosis but also angio-genesis, and might be upstream of VEGF in cancer cells. Theexpression of VEGF was also significantly decreased by down-regulation of PCA-1/ALKBH3 in vivo. And then the number of

Table 1. Comparison of clinicopathologicalcharacteristics according to tumor PCA-1/ALKBH3 expression

CharacteristicsPCA-1 low(n ¼ 59)

PCA-1 high(n ¼ 57) P value

Age, y 66 (42–81) 66 (33–80) 0.6480GenderMale 33 (55.9%) 34 (59.7%) 0.6854Female 26 (44.1%) 23 (40.4%)

Histopathological gradingG1 17 (28.8%) 17 (29.8%) 0.6156G2 34 (57.6%) 31 (54.4%)G3 5 (8.5%) 8 (14.0%)G4 3 (5.1%) 1 (1.8%)

pT categorypT1/pT2/pT3 53 (89.8%) 40 (70.2%) 0.0079pT4 6 (10.2%) 17 (29.8%)

pN categorypN0 28 (47.5%) 19 (40.4%) 0.1214pN1 31 (52.5%) 38 (59.6%)

pM categorypM0 57 (96.6%) 50 (87.7%) 0.0735pM1 2 (3.4%) 7 (12.3%)

UICC stageIA/IB 14 (23.7%) 7 (12.3%) 0.0144IIA/IIB 38 (64.4%) 29 (50.9%)III 5 (8.5%) 14 (24.6%)IV 2 (3.4%) 7 (12.3%)

VEGF intensityHigh 22 (37.3%) 42 (73.7%) <0.0001Low 37 (62.7%) 15 (26.3%)

Abbreviation: UICC, International Union Against Cancer.






microvessels in tumors treated with PCA-1/ALKBH3 siRNAwas significantly lower compare to that in tumors treated withcontrol RNA. Although PCA-1/ALKBH3 knockdown led toapoptosis in 29% of PANC-1 cells in vitro, it led to about50% to 70% of tumor reduction compared with controlsin vivo subcutaneous and orthotopic models. Taken together,data suggested that apoptosis induction and anti-angiogenesisinduced by PCA-1/ALKBH3 knockdown might synergisticallyexert antitumor effect in vivo.

In our study, PCA-1/ALKBH3 knockdown induced not onlyapoptosis but also reduction of VEGF in 2 human pancreaticcancer cell lines. However, the relations of PCA-1/ALKBH3 andangiogenesis are still unknown. Therefore, further fundamen-tal studies are clearly required. Collectively PCA-1/ALKBH3might be a potential target against pancreatic cancer, becauseblockade of PCA-1/ALKBH3 induced both apoptosis and anti-angiogenesis.

Finally we examined clinical importance of PCA-1/ALKBH3 in actual human pancreatic cancer, and found thatPCA-1/ALKBH3 expression was abundant in most humanpancreatic cancer tissues and was not seen in noncancertissues. The expression of PCA-1/ALKBH3 may be helpfulfor the early diagnosis of pancreatic cancer. The tumorPCA-1/ALKBH3 expression in human pancreatic cancersignificantly correlated with advanced tumor status andpathological stage, whereas it did not relate to lymph nodeand distant metastasis. These clinical data suggested thattumor PCA-1/ALKBH3 might be an important factor ofproliferation and invasion rather than that of metastasis inpancreatic cancer. Therefore, further study to evaluate theserum level of PCA-1/ALKBH3 in pancreatic cancer patientsmay be warranted. In addition, patients with high positivityof PCA-1/ALKBH3 expression have significantly poor prog-nosis in comparison with low positivity of PCA-1/ALKBH3expression. Furthermore, we showed correlation betweenpositivity of PCA-1/ALKBH3 and intensity of VEGF expres-sion in human cancer. These data further emphasized that

PCA-1/ALKBH3 may be a promising therapeutic targetagainst human pancreatic cancer.

In conclusion, we have shown for the first time that PCA-1/ALKBH3 participates in apoptotic resistance of cancer cells,proliferation of tumor and tumor angiogenesis in pancreaticcancer. An inverse correlation has been also observed betweenPCA-1/ALKBH3 expression and prognosis in pancreatic cancerpatients. This study may provide the rationale of developing anovel cancer therapy targeting PCA-1/ALKBH3 for this fatalmalignant disease.

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

Authors' ContributionsConception and design: I. Yamato, M. Sho, K. Shimada, N. Konishi, K.Tsujikawa, Y. NakajimaDevelopment of methodology: I. Yamato, M. Sho, K. Shimada, Y. Ueda, N.Konishi, K. TsujikawaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): I. Yamato, M. Sho, K. Shimada, K. Hotta, Y. Ueda, S.Yasuda, N. Konishi, K. TsujikawaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): I. Yamato, M. Sho, K. Hotta, K. TsujikawaWriting, review, and/or revision of the manuscript: I. Yamato, M. Sho, N.Konishi, K. TsujikawaAdministrative, technical, or material support (i.e., reporting or orga-nizingdata, constructingdatabases):M. Sho, S. Yasuda, N. Shigi, K. TsujikawaStudy supervision: M. Sho, K. Shimada, N. Konishi, K. Tsujikawa, Y. Nakajima

Grant SupportsThis study was supported by the following grants: Grants-in-Aid for Scientific

Research from the Ministry of Education, Culture, Sports, Science and Tech-nology of Japan (M. Sho) and Kansai Biomedical Cluster project promoted by theKnowledge Cluster Initiative (stage-II) of the Ministry of Education, Culture,Sports, Science and Technology, Japan, and the Program for Promotion ofFundamental Studies in Health Sciences of the National Institute of BiomedicalInnovation (K. Tsujikawa).

The costs of publication of this article were defrayed in part 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.

Received January 30, 2012; revised July 3, 2012; accepted July 17, 2012;published OnlineFirst July 23, 2012.

References1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Murray T, et al. Cancer

statistics, 2008. CA Cancer J Clin 2008;58:71–96.2. Cunningham D, Chau I, Stocken DD, Valle JW, Smith D, Steward W,

et al. Phase III randomized comparison of gemcitabine versus gemci-tabine plus capecitabine in patients with advanced pancreatic cancer.J Clin Oncol 2009;27:5513–8.

3. HerrmannR, BodokyG, Ruhstaller T, Glimelius B, Bajetta E, Schuller J,et al. Gemcitabine plus capecitabine compared with gemcitabinealone in advanced pancreatic cancer: a randomized, multicenter,phase III trial of the Swiss Group for Clinical Cancer Research andthe Central European Cooperative Oncology Group. J Clin Oncol2007;25:2212–7.

4. Ferrone CR, Kattan MW, Tomlinson JS, Thayer SP, Brennan MF,Warshaw AL. Validation of a postresection pancreatic adenocarcino-ma nomogram for disease-specific survival. J Clin Oncol 2005;23:7529–35.

5. WagnerM,Redaelli C,LietzM,SeilerCA,FriessH,BuchlerMW.Curativeresection is the single most important factor determining outcome inpatients with pancreatic adenocarcinoma. Br J Surg 2004;91:586–94.

6. Maheshwari V, Moser AJ. Current management of locally advancedpancreatic cancer. Nat Clin Pract Gastroenterol Hepatol 2005;2:356–64.

7. Martin SJ, Green DR. Apoptosis and cancer: the failure of controlson cell death and cell survival. Crit Rev Oncol Hematol 1995;18:137–53.

8. Mangray S, King TC. Molecular pathobiology of pancreatic adenocar-cinoma. Front Biosci 1998;3:D1148–60.

9. Okada H, Mak TW. Pathways of apoptotic and non-apoptotic death intumour cells. Nat Rev Cancer 2004;4:592–603.

10. Konishi N, Nakamura M, Ishida E, Shimada K, Mitsui E, Yoshikawa R,et al. High expression of a new marker PCA-1 in human prostatecarcinoma. Clin Cancer Res 2005;11:5090–7.

11. Tasaki M, Shimada K, Kimura H, Tsujikawa K, Konishi N. ALKBH3, ahuman AlkB homologue, contributes to cell survival in human non-small-cell lung cancer. Br J Cancer 2011;104:700–6.

12. Shimada K, Nakamura M, Anai S, De Velasco M, Tanaka M, TsujikawaK, et al. A novel human AlkB homologue, ALKBH8, contributes tohuman bladder cancer progression. Cancer Res 2009;69:3157–64.

13. Aas PA, Otterlei M, Falnes PO, Vagbo CB, Skorpen F, Akbari M, et al.Human and bacterial oxidative demethylases repair alkylation damagein both RNA and DNA. Nature 2003;421:859–63.

14. Falnes PO, Johansen RF, Seeberg E. AlkB-mediated oxidativedemethylation reverses DNA damage in Escherichia coli. Nature2002;419:178–82.

Yamato et al.





15. Chen B, Liu H, Sun X, Yang CG. Mechanistic insight into the recog-nition of single-stranded and double-stranded DNA substrates byABH2 and ABH3. Mol Biosyst 2010;6:2143–9.

16. Kurowski MA, Bhagwat AS, Papaj G, Bujnicki JM. Phylogenomicidentification of five new human homologs of the DNA repair enzymeAlkB. BMC Genomics 2003;4:48.

17. Gerken T, Girard CA, Tung YC, Webby CJ, Saudek V, Hewitson KS,et al. The obesity-associated FTO gene encodes a 2-oxoglutarate-dependent nucleic acid demethylase. Science 2007;318:1469–72.

18. Tsujikawa K, Koike K, Kitae K, Shinkawa A, Arima H, Suzuki T, et al.Expression and sub-cellular localization of human ABH family mole-cules. J Cell Mol Med 2007;11:1105–16.

19. Drablos F, Feyzi E, Aas PA, Vaagbo CB, Kavli B, Bratlie MS, et al.Alkylation damage in DNA and RNA–repair mechanisms and medicalsignificance. DNA Repair (Amst) 2004;3:1389–407.

20. Ougland R, Zhang CM, Liiv A, Johansen RF, Seeberg E, Hou YM, et al.AlkB restores the biological function ofmRNA and tRNA inactivated bychemical methylation. Mol Cell 2004;16:107–16.

21. Hausinger RP. FeII/alpha-ketoglutarate-dependent hydroxylases andrelated enzymes. Crit Rev Biochem Mol Biol 2004;39:21–68.

22. Yu B, EdstromWC, Benach J, Hamuro Y, Weber PC, Gibney BR, et al.Crystal structures of catalytic complexes of the oxidative DNA/RNArepair enzyme AlkB. Nature 2006;439:879–84.

23. SundheimO, VagboCB, BjorasM, SousaMM, Talstad V, Aas PA, et al.Human ABH3 structure and key residues for oxidative demethylationto reverse DNA/RNA damage. EMBO J 2006;25:3389–97.

24. Shimada K, NakamuraM, Ishida E, Higuchi T, Yamamoto H, TsujikawaK, et al. Prostate cancer antigen-1 contributes to cell survival andinvasion thoughdiscoidin receptor 1 in humanprostate cancer.CancerSci 2008;99:39–45.

25. Dinglay S, Trewick SC, Lindahl T, Sedgwick B. Defective processing ofmethylated single-stranded DNA by E. coli AlkB mutants. Genes Dev2000;14:2097–105.

26. Yamato I, ShoM, Nomi T, Akahori T, Shimada K, Hotta K, et al. Clinicalimportance of B7-H3 expression in human pancreatic cancer. BrJ Cancer 2009;101:1709–16.

27. Sobin L, Wittekind C. UICC-TNM classification of malignant tumours.6th ed. New York: Wiley-Liss; 2002.

28. Hecht SS. Progress and challenges in selected areas of tobaccocarcinogenesis. Chem Res Toxicol 2008;21:160–71.

29. Rydberg B, Lindahl T. Nonenzymatic methylation of DNA by theintracellular methyl group donor S-adenosyl-L-methionine is a poten-tially mutagenic reaction. EMBO J 1982;1:211–6.

30. Bestor T, Laudano A, Mattaliano R, Ingram V. Cloning and sequencingof a cDNA encoding DNA methyltransferase of mouse cells. Thecarboxyl-terminal domain of the mammalian enzymes is related tobacterial restriction methyltransferases. J Mol Biol 1988;203:971–83.

31. Jin B, Yao B, Li JL, Fields CR, Delmas AL, Liu C, et al. DNMT1 andDNMT3B modulate distinct polycomb-mediated histone modifica-tions in colon cancer. Cancer Res 2009;69:7412–21.

32. Gopalakrishnan S, Van Emburgh BO, Robertson KD. DNAmethylationin development and human disease. Mutat Res 2008;647:30–8.

33. Robertson KD. DNA methylation and human disease. Nat Rev Genet2005;6:597–610.

34. Nieminuszczy J, Grzesiuk E. Bacterial DNA repair genes and theireukaryotic homologues: 3. AlkB dioxygenase and Ada methyltrans-ferase in the direct repair of alkylated DNA. Acta Biochim Pol2007;54:459–68.

35. Lee DH, Jin SG, Cai S, Chen Y, Pfeifer GP, O'Connor TR. Repair ofmethylation damage in DNA and RNA by mammalian AlkB homolo-gues. J Biol Chem 2005;280:39448–59.

36. Monsen VT, Sundheim O, Aas PA, Westbye MP, Sousa MM, Slup-phaug G, et al. Divergent ss-hairpins determine double-strand versussingle-strand substrate recognition of human AlkB-homologues 2 and3. Nucleic Acids Res 2010;38:6447–55.

37. Yi C, Yang CG, He C. A non-heme iron-mediated chemical demeth-ylation in DNA and RNA. Acc Chem Res 2009;42:519–29.

38. Choi SY, Jang JH, KimKR. Analysis of differentially expressedgenes inhuman rectal carcinoma using suppression subtractive hybridization.Clin Exp Med 2011.

39. Neta G, Brenner Annexin V, Sturgis EM, Pfeiffer RM, Hutchinson AA,Aschebrook-Kilfoy B, et al. Common genetic variants related to geno-mic integrity and risk of papillary thyroid cancer. Carcinogenesis2011;32:1231–7.

40. WuSS, XuW, Liu S, Chen B,Wang XL,Wang Y, et al. Down-regulationof ALKBH2 increases cisplatin sensitivity in H1299 lung cancer cells.Acta Pharmacol Sin 2011;32:393–8.

41. GaoW, Li L, Xu P, Fang J, XiaoS, ChenS. Frequent down-regulation ofhABH2 in gastric cancer and its involvement in growth of cancer cells.J Gastroenterol Hepatol 2011;26:577–84.

42. Folkman J, Merler E, Abernathy C, Williams G. Isolation of a tumorfactor responsible for angiogenesis. J Exp Med 1971;133:275–88.

43. Gale NW, Thurston G, Davis S, Wiegand SJ, Holash J, Rudge JS, et al.Complementary and coordinated roles of theVEGFsandangiopoietinsduring normal and pathologic vascular formation. Cold Spring HarbSymp Quant Biol 2002;67:267–73.

44. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases.Nature 2000;407:249–57.

45. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and otherdisease. Nat Med 1995;1:27–31.

46. KuwaharaK, Sasaki T, KuwadaY,MurakamiM, Yamasaki S, ChayamaK. Expressions of angiogenic factors in pancreatic ductal carcinoma: acorrelative study with clinicopathologic parameters and patient sur-vival. Pancreas 2003;26:344–9.

47. Ikeda N, Adachi M, Taki T, Huang C, Hashida H, Takabayashi A, et al.Prognostic significance of angiogenesis in human pancreatic cancer.Br J Cancer 1999;79:1553–63.

48. Khan AW, Dhillon AP, Hutchins R, Abraham A, Shah SR, Snooks S,et al. Prognostic significance of intratumoural microvessel density(IMD) in resected pancreatic and ampullary cancers to standardhistopathological variables and survival. Eur J Surg Oncol 2002;28:637–44.

49. Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J,Heim W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucov-orin formetastatic colorectal cancer. N Engl JMed 2004;350:2335–42.

50. Sandler A, Gray R, Perry MC, Brahmer J, Schiller JH, Dowlati A, et al.Paclitaxel-carboplatin alone or with bevacizumab for non-small-celllung cancer. N Engl J Med 2006;355:2542–50.






2012;72:4829-4839. Published OnlineFirst July 23, 2012.Cancer Res Ichiro Yamato, Masayuki Sho, Keiji Shimada, et al. Apoptotic Resistance and AngiogenesisPCA-1/ALKBH3 Contributes to Pancreatic Cancer by Supporting

Updated version

10.1158/0008-5472.CAN-12-0328doi:

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/72/18/4829.full#ref-list-1

This article cites 48 articles, 11 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/72/18/4829.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles 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://cancerres.aacrjournals.org/content/72/18/4829To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-12-0328

http://cancerres.aacrjournals.org/content/72/18/4829.full#ref-list-1

http://cancerres.aacrjournals.org/content/72/18/4829.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/72/18/4829


Documents

PCA-1/ALKBH3 Contributes to Pancreatic Cancer by Supporting … · pancreatic cancer and therefore can be potential therapeutic target. Using RNA interference method in vitro and