Cell Cycle and Senescence

A Novel Regulatory Mechanism of Pim-3 Kinase Stability andIts Involvement in Pancreatic Cancer Progression

Fei Zhang1, Bin Liu1, Zhen Wang1, Xian-Jun Yu2, Quan-Xing Ni2, Wen-Tao Yang3,4,Naofumi Mukaida5, and Ying-Yi Li1,5

AbstractTranslationally controlled tumor protein (TCTP/TPT1) was identified from a yeast 2-hybrid screen and shown

to interact with Pim-3, a member of the proto-oncogene Pim family with serine/threonine kinase activity. TCTPwas aberrantly expressed in human pancreatic cancer cells and malignant ductal epithelial cells, but not in normalpancreatic duct epithelial cells adjacent to tumor foci of human pancreatic cancer tissue. Moreover, TCTPcolocalized with Pim-3 both in human pancreatic cancer cells and in clinical tissues. Mapping studies revealed thatthe interaction between Pim-3 and TCTP occurred through the C-terminal region of Pim-3 andN-terminal regionof TCTP. Although Pim-3 had no effect on TCTP expression or phosphorylation, overexpression of TCTPincreased the amount of Pim-3 in a dose-dependent manner. Interestingly, RNAi-mediated ablation of TCTPexpression reduced Pim-3 protein but not mRNA, through a mechanism involving the ubiquitin–proteasomedegradation system. As a consequence of Pim-3 instability and subsequent degradation, tumor growth in vitro andin vivo was inhibited by arresting cell-cycle progression and enhancing apoptosis. Furthermore, TCTP and Pim-3expression were significantly correlated in pancreatic adenocarcinoma specimens, and patients with highlyexpressed TCTP and Pim-3 presented with a more advanced tumor stage. These observations indicate thatTCTP enhances Pim-3 stability to simultaneously promote and prevent cell-cycle progression and apoptosis,respectively. Hence, TCTP and Pim-3 serve a pivotal role in human pancreatic cancer with important ramificationsfor clinical diagnostic and therapeutic implications.

Implications: The present study provides a new idea and experimental evidence for recognizing TCTP/Pim-3pathway as a target for therapy in human pancreatic cancer. Mol Cancer Res; 11(12); 1508–20. �2013 AACR.

IntroductionPim-3, a member of the proto-oncogene Pim family

with serine/threonine kinase activity, was originally iden-tified as a depolarization-induced gene, KID-1, in the ratpheochromocytoma cell line PC12 (1). Later, KID-1 wasrenamed Pim-3 because of its high sequence similaritywith Pim family proteins, which belong to the group ofcalcium/calmodulin-regulated kinase (1). Subsequently,Deneen and colleagues demonstrated that Pim-3 gene

transcription was enhanced in EWS/ETS-induced malig-nant transformation of NIH3T3 cells (2), suggesting theinvolvement of Pim-3 in tumorigenesis. In line with theseobservations, we demonstrated that Pim-3 expression wasenhanced in malignant lesions, but not normal tissues ofendoderm-derived organs such as the liver (3), pancreas(4), colon (5), and stomach (6). Hepatocellular carcinomadevelopment was accelerated in mice expressing the Pim-3transgene selectively in liver, when these mice were treatedwith a hepatocarcinogen (7). We observed that Pim-3 caninactivate Bad in human pancreas and colon carcinomacell lines by phosphorylating Ser112, but not Ser136, andultimately promoting their survival (4, 5) as observed forPim-1 and Pim-2 (8, 9). Moreover, it has been reportedthat Pim-3 can promote cell-cycle progression by modu-lating the functions of molecules that regulate cell-cycleprogression and augment protein synthesis through theregulation of PGC-1a, eventually contributing to carci-nogenesis (10). Baines and colleagues recently reportedthat Pim-3 suppression can sensitize pancreatic cancercells to gemcitabine (11). We also demonstrated thatPim-3 can promote tumor growth and angiogenesis bystimulating the VEGF pathway (12). Furthermore, Pim-3modulates Myc activity to promote tumorigenesis (13).

Authors' Affiliations: 1Cancer Research Institute; 2Department of Pan-creas and Hepatobiliary, Pancreatic Cancer Institute; 3Department ofPathology, Fudan University Shanghai Cancer Center; 4Department ofOncology, Shanghai Medical College, Fudan University, Shanghai, China;and 5Division of Molecular Bioregulation, Cancer MicroenvironmentResearch Program, Cancer Research Institute, Kanazawa University,Kanazawa, Japan

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

Corresponding Author: Professor Ying-Yi Li, Room 1216, 2nd Building,Cancer Research Institute, Fudan University Shanghai Cancer Center, 270DongAn Road, Shanghai 200032, China. Phone: 86-21-64175590-5220;Fax: 86-21-64172585; E-mail: liyingyi@fudan.edu.cn

doi: 10.1158/1541-7786.MCR-13-0389

�2013 American Association for Cancer Research.

MolecularCancer

Research

Mol Cancer Res; 11(12) December 20131508

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

Thus, Pim-3 is a key player in tumorigenesis, and there-fore, an ideal target for cancer therapy.The translationally controlled tumor protein (TCTP) is a

highly conserved hydrophilic protein (14) that has beenidentified in a wide range of eukaryotic organisms, includingfungi, yeast, insects, plants, andmammals (15). TCTP is alsoknown as IgE-dependent histamine-releasing factor (HRF),fortilin, P21, P23, and TPT-1 (16–18). This protein wasnamed TCTP because its mRNAwas found to be controlledat the translational level (19). Although TCTP is foundubiquitously in tissues and cell types, its expression levelsvary depending on the tissue type, growth, stress factors, andcytotoxic signals (20–22). A series of recent reports provedthat TCTP plays important roles in a number of cellphysiological events in cancer, cell proliferation, apoptosisregulation, stress response, gene regulation, heat shockresponse, and allergic response (23). TCTP can also interactwithmany cellular proteins, including translation elongationfactors eEF1A and eEF-B-b, tubulin, actin, myeloid cellleukemia protein-1 (MCL1), Bcl-XL, p53, and Na- and K-ATPase (24, 25). However, the roles of TCTP in tumor-igenesis remain largely unknown.Considering the critical role of Pim-3 in tumor develop-

ment and progression, defining regulatory mechanisms ofPim-3 signaling networks is important. In this study, toidentify potential novel regulators of Pim-3, we performedyeast 2-hybrid screening using human HeLa matchmakercDNA library. We observed that TCTP specifically interactswith Pim-3 and enhances Pim-3 protein stability by blockingthe ubiquitin-proteasome-mediated degradation of Pim-3protein, thereby promoting tumor growth in vitro and in vivo.

Materials and MethodsCell culture and antibodiesHuman pancreatic carcinoma cell lines, PCI35 and

PCI55, were gifts from Prof. Mukaida Naofumi of Kana-zawa University (Kanazawa, Japan). Human pancreaticcarcinoma cell lines SW1990, MiaPaca-2, PANC-1, andBxPC-3 and the human embryonic kidney (HEK) cell line293T were purchased from the American Type CultureCollection. Among them, SW1990, MiaPaca-2, andPANC-1 cells were cultured in RPMI 1640 supplementedwith 10% FBS (Biowest, Inc.). BxPC-3 and 293T cells werecultured in Dulbecco's Modified Eagle Medium supple-mented with 10% FBS at 37�C in a humidified 5% CO2incubator. The authenticity of all the cell lines was confirmedby determiningDNAprofiling of short tandem repeat (STR)whereas mycoplasma contamination was excluded with thehelp of Amelogenin (Beijing Microread Genetics Co., Ltd.)The antibodies used in this study are described in Supple-mentary Materials and Methods.

Yeast 2-hybrid screeningHuman Pim-3 cDNA fragment (121-326aa) was cloned

into pGBKT7 vector and used to screen a pACT2-humanHeLa matchmaker cDNA library in a yeast 2-hybrid system(Clontech), because we did not detect Pim-3 protein inHeLacells (4). b-Galactosidase activities were measured using o-

nitrophenyl-galactoside as a substrate. Clones activating theb-galactosidase reporter gene were sequenced and analyzed.

Generation of expression vectorsAn expression plasmid for TCTP tagged with the His

epitope at the COOH- terminus was constructed to detectTCTP protein by the anti-His antibody. A TCTP cDNAfragment, which has an XhoI restriction site in place of thestop codon, was generated by PCR using TCTP cDNAcloned in the pACT2 plasmid as the template and 2 primers(Supplementary Table S1). The amplified fragments weredigested with HindIII and XhoI (Takara) and inserted intoHindIII and XhoI sites of the pcDNA4 vector (Invitrogen).The Pim-3 expression vector (pcDNA4-Pim-3) and thePim-3-shRNA vector (pSilencer-Pim-3-shRNA) were con-structed as previously described (4).Three deletion mutants of TCTP and 2 deletion mutants

of Pim-3 were isolated by PCR using a combination ofprimers (Supplementary Table S1) for His-tagged con-structs. PCR products spanning each fragment were clonedinto the EcoRI and XhoI sites of the pcDNA4 vector, andnamed DTCTP-1 (Met1-Gly69), DTCTP-2 (Val70-Ala119),DTCTP-3 (Glu120-Cys172), DPim-3A(met1-pro126), andDPim-3B (leu121-leu326), respectively. All mutated nucleo-tides were confirmed by sequencing.

Co-immunoprecipitation and immunoblottingHEK293T cells were cotransfected with the Pim-3 expres-

sion and pcDNA4-TCTP-His vectors using Lipofectamine2000 (Invitrogen) according to the manufacturer's protocol.Forty-eight hours after transfection, the cells were collectedand solubilized with 1 mL of NP-40 lysis buffer [50 mmol/LTris-HCl, 150 mmol/L NaCl, 1% NP-40, 1 mmol/LEDTA, pH 7.4, with complete protein inhibitor cocktail(Roche)]. Precleared cell lysates were incubated with 2 mgrabbit anti-Pim-3 antibody or mouse anti-His antibodyovernight at 4�C and precipitated with 20 mL protein G-Sepharose 4 Fast flow (GEHealthcare Bio-sciences AB) for 2to 4 hours at 4�C. The beads were washed 3 times with celllysis buffer. Materials bound to the beads were eluted withSDS-PAGE loading buffer containing 1% b-mercaptoetha-nol, boiled for 5 minutes, and separated on 12% SDS-polyacrylamide gels and transferred to polyvinylidenedifluor-ide membranes (Millipore). Immunoblotting was performedas previously described (4).

Immunofluorescence analysisPCI55 cells were cultured for 48 hours, washed and fixed

with 4% paraformaldehyde/PBS, and hybridized with acombination of goat polyclonal anti-Pim-3 antibodies(1:100; Santa Cruz) and rabbit monoclonal anti-TCTPantibodies (1:100; Epitomics). Next, the cells were sequen-tially incubated with Alexa Fluor 594 donkey anti-goatimmunoglobulin G (IgG) and Alexa Fluor 488-conjugatedgoat anti-rabbit IgG antibodies. The signals were visualizedusing immunofluorescence confocal microscopy (Leica).Immunofluorescence analysis was performed similarly onhuman pancreatic cancer tissues.

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1509

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

Real-time quantitative RT-PCRTotal RNA was extracted using the Trizol LS reagent

(Invitrogen). mRNA was reverse transcribed using theSuperScript First-Strand Synthesis System (Invitrogen).Real-time PCR was performed using the Applied Biosys-tems HT7900 PCR system with 2� QuantiFast SYBRGreen PCR Master Mix (Qiagen), 0.2 mmol/L primers(Supplementary Table S2), and <100 ng cDNA in a 25-mL reaction mixture. Relative expression of target geneswas analyzed by the DDCt method. Results are expressedas means � SD.

Knockdown of TCTP expressionThe selected short interfering RNA target sequence

in TCTP (50-AAGGTACCGAAAGCACAGT-30 corre-sponded to 179–197 residues) and nonspecific controlshort interfering RNA duplexes (50-UUCUCCGAAC-GUGUCACGUTT-30) were designed and synthesized byGenePharma (Shanghai Co., Ltd.). For transient knock-down of TCTP expression, PCI55 cells were transientlytransfected with the resultant siRNA using Lipofectamine2000. The small hairpin RNA (shRNA)-encoding oligo-nucleotides for TCTP and scramble were prepared bySangon (Shanghai). The annealed shRNA were insertedinto the AgeI and EcoRI sites of the lentiviral plasmidpLKO.1-TRC cloning vector (Addgene). For stableknockdown of TCTP, HEK293T cells were plated in75-cm2 culture flasks and transfected with 10 mg TCTPshRNA or scramble shRNA lentiviral vectors. The medi-um was changed the next day and viral supernatant washarvested 48 hours later. All viral containing medium wascollected, passed through 0.45-mm syringe filters. PCI55,Miapaca-2, and SW1990 cells were incubated with thelentivirus supernatant for 24 hours and selected withpuromycin (4 mg/mL).

Cell cycle, cell apoptosis, and cell viabilityFlow cytometric analysis was conducted to examine cell

cycle and apoptosis with the help of propidium iodine(Invitrogen) and human Annexin V-FITC Kit (Invitrogen),respectively, according to the manufacturer's protocol. Cellviability was determined at 0, 24, 48, 72, 96, and 120 hoursusing the Cell Counting Kit-8 reagent (Dojindo) accordingto the manufacturer's protocol. All observations were repro-duced at least 3 times in independent experiments.

Mouse xenograftsFemale Balb/c nudemice (6–8weeks of age, weighing 18–

20 g, and specific pathogen free) were obtained fromShanghai SLAC Laboratory Animals (Shanghai, China).Before the experiment, mice were divided into 4 groups(Mipaca-2-scramble shRNA, Miapaca-2-TCTP shRNA,SW1990-scramble shRNA, and SW1990-TCTP shRNA).Each cancer cell line (5 � 106/site) was injected subcuta-neously into the right flank of a nude mouse. After estab-lishment of the nudemice xenograft model, tumor sizes weremeasured every 3 to 4 days using micrometer calipers.Tumor volumes were calculated using the following formu-

la: volume ¼ 1/2a � b2, where a and b represent the largerand smaller tumor diameters, respectively. Tumor growthwas followed for 28 days from the first injection. All animalexperiments were performed in compliance with the Guide-lines for the Care and Use of Laboratory Animals of FudanUniversity. The protocol was approved by the Committeeon the Ethics of Animal Experiments of Fudan University[permit number, SYXK(Hu)2009-0082].

Patient and tissue samplesThe study included 148 patients who underwent surgery

at Fudan University Shanghai Cancer Center from 2004 to2012 after histological verification of pancreatic ductaladenocarcinoma. None of these patients received preoper-ative chemotherapy or radiotherapy. The patients providedwritten consent for the use of tumor tissue for clinicalresearch, and the Fudan University Shanghai Cancer CenterEthical Committee approved the research protocol. Detaileddescription of patients and tissue samples is provided inSupplementary Materials and Methods.

Immunohistochemical analysis of human pancreaticcancer tissuesFollowing deparaffinization and quenching of endogenous

peroxidase, sections were incubated with 1% bovine serumalbumin (BSA) in PBS. Subsequently, the slides were treatedwith rabbit anti-TCTP (1:500) and goat anti-Pim-3 (1:25)antibodies followed by incubation with donkey anti-goat IgGand goat anti-rabbit IgG antibodies, respectively. Pim-3 andTCTP immunoreactivities were visualized using the GTvision DAB kit (GeneTech). The slides were counterstainedwith ChemMate Hematoxylin (DakoCytomation) andmounted and observed under a microscope (Olympus). Theproportion of Pim-3- or TCTP-positive cells in humanpancreatic carcinoma tissues were evaluated by 2 independentpathologists without a prior knowledge on the clinical infor-mation. The scoring of TCTP and Pim-3 was performed asdescribed in Supplementary Materials and Methods.

Statistical analysisThe x2 test or Fisher exact probability test was used to

compare clinicopathological features of the 148 patientswith TCTP and Pim-3 expression. Correlation betweenTCTP and Pim-3 expression was evaluated using Spearmancorrelation analysis. Statistical analysis was performed withSPSS statistical software (IBMSPSS Statistics 20). Data werereported as the means� SD when appropriate and P < 0.05was considered statistically significant.

ResultsInteraction between Pim-3 and TCTPWe initially screened a cDNA library to identify the gene

products that can modulate Pim-3 activity as well as serve asits substrate. Using the yeast 2-hybrid system with Pim-3 asbait, 6 candidate genes were identified (data not shown).Among them, we characterized human TCTP protein(NM_003295.2) at first. To validate the specificity of theinteraction between Pim-3 and TCTP, His-tagged TCTP

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1510

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

IP: Pim-3

A B C

D E

G H

I

K

J

F

Pim-3IgG His TCTPInputIP

IgG IgG

Pim-3

His

Pim-3

TCTP

1 2 3 4 1 2 3 4

TCTP Pim-3

DAPI

P < 0.01

Merged

20 μmol/L

SW1990 PC135 PC155 PANC-1 BxPC-3 MiaPaca-2

TCTP

Pim-3

Pim-3 N

N

N

TCTPPim-3

β-Actin

1.2

1

0.8

0.6

0.4

0.2

0

SW19

90

PC13

5

PC15

5

PANC-1

BxPC-3

MiaPac

a-2

1

0.8

0.6

0.4

0.2

00 0.2 0.4 0.6 0.8 1 1.2

TCTP protein (normalized to actin)

Pim

-3 p

rote

in

(norm

aliz

ed to A

ctin)

TC

TP

and P

im-3

(Pro

tein

/actin p

rote

in)

IP: Pim-3

IP: TCTP

Pim-31aa

1aa 69aa

70aa 119aa

120aa 172aa

326aa1aa

1aa 126aa

121aa 326aa

172aa

C

C

C

C

C

C

C

Pim-3

Δ TCTP-1

Δ Pim-3AΔ Pim-3B

Δ Pim-3A

Δ Pim-3B

Pim-3

Pim-3

ΔPim-3B

TCTP

TCTP

121-126

Pim-3

ΔPim-3A ΔPim-3B

Δ TCTP-1

Δ TCTP-1 Δ TCTP-2 Δ TCTP-3

1-69aa 70-119aa 120-172aa

1-120aa 127-326aa

Δ TCTP-2Δ TCTP-3

TCTP

TCTP

TCTP ++ − −

−−− −

+

+

+

+

+ + + ++ − − −− + − −− − + −− − − +

TCTP N

Δ TCTP-1 N

Δ TCTP-2 N

Δ TCTP-3 N

Figure 1. Association of TCTP with Pim-3 in human pancreatic cancer cell lines. A, co-immunoprecipitation of TCTP and Pim-3 in transiently transfectedHEK293Tcells. Cell lysateswereobtained fromHEK293Tcells transiently transfectedwithpcDNA4-Pim-3 andpcDNA4-TCTP-His. The resultant lysatesweresubjected to immunoprecipitation and immunoblotting. B, co-immunoprecipitation of endogenous TCTP and Pim-3 in PCI55 cells. Cell lysates wereobtained from PCI55 cells. The resultant lysates were subjected to immunoprecipitation and immunoblotting. One tenth of the cell lysates from PCI55 weresubjected to immunoblotting with anti-TCTP or anti-Pim-3 antibodies as positive controls. C, colocalization of TCTP and Pim-3 in human pancreaticcancer cells. PCI55 cells were immunostained with a combination of anti-TCTP and anti-Pim-3 antibodies as described in Materials and Methods. Thefluorescent images were digitally merged. Yellow coloration in overlay panels indicates colocalization of Pim-3 and TCTP. Nuclei were counterstained with40,6-diamidino-2-phenylindole. D, TCTP and Pim-3 expression in human pancreatic cancer cell lines was detected by Western blotting. b-Actin was used asan internal control. E, the expression levels of TCTP, Pim-3, and b-actin were quantified using NIH ImageJ software. TCTP/b-actin and Pim-3/b-actin ratioswere calculated for each cell line. F, correlation between TCTP and Pim-3 expression in human pancreatic cancer cell lines, with a linear regressionline and Spearman correlation significance (Spearman correlation coefficient¼ 1, P < 0.01). G, schematic representation of cDNA constructs for each TCTPdeletionmutant. H, cell lysateswere obtained fromHEK293T cells transiently cotransfectedwith Pim-3 andTCTPdeletionmutants. The resultant lysateswereimmunoprecipitated and immunoblotted. I, schematic representation of cDNA constructs for each Pim-3 deletion mutant. J, cell lysates were obtained fromHEK293T cells transiently cotransfected with TCTP and Pim-3 deletion mutants. The resultant lysates were immunoprecipitated and immunoblotted.K, schematic representation of the presumed interaction between Pim-3 and TCTP proteins.

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1511

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

protein was coexpressed with Pim-3 in HEK293T cells.Immunoprecipitation with anti-Pim-3 antibodies but notcontrol IgG, coprecipitated His-tagged TCTP (Fig. 1A,compare lanes 1 and 2). Moreover, immunoprecipitationwith anti-His antibodies but not control IgG also specificallycoprecipitated Pim-3 protein (Fig. 1A, lanes 3 and 4). Next,we examined the interaction between Pim-3 and TCTP inPCI55 cells at endogenous levels. Immunoprecipitationwith anti-TCTP antibodies, but not control IgG, copreci-pitated endogenous Pim-3 (Fig. 1B, compare lanes 2 and 4).In a reciprocal co-IP experiment, the TCTP protein wasfound to be present in the immune complexes pulled downby the anti-Pim-3 antibodies but not by the control IgG (Fig.1B, compare lanes 3 and 4). Furthermore, immunofluores-cence analysis showed that TCTP and Pim-3 were predom-inantly localized in the cytoplasm and colocalized in humanpancreatic cancer cell lines (Fig. 1C). These observationsindicate a physical interaction between Pim-3 and TCTP,and prompted us to examine TCTP expression in humanpancreatic cancer cell lines.We detectedTCTPmRNA (datanot shown) and protein expression in all 6 human pancreaticcancer cell lines that we examined (Fig. 1D and E). More-over, TCTP expression correlated with Pim-3 expression inthese cell lines (Spearman correlation coefficient, 1.0; P <0.01; Fig. 1F). To map the region necessary for the inter-

action between Pim-3 and TCTP, we constructed trunca-tion vectors of TCTP and Pim-3. As shown in Fig. 1G andH, Pim-3 was coprecipitated with DTCTP-1 deletionmutants but not with the other deletion mutant(DTCTP-2 and DTCTP-3). Moreover, the Pim-3 mutantlacking the C-terminal domain was not coprecipitated byTCTP, whereas the Pim-3 mutant lacking N-terminaldomain was (Fig. 1I and J). Taken together, these resultsindicate that the interaction between Pim-3 and TCTP isimmediately mediated through the C-terminal region ofPim-3 and N-terminal region of TCTP (Fig. 1K).

Pim-3 has no effect on TCTP expression orphosphorylationWe previously observed that Pim-3 can prevent cell apo-

ptosis and eventually lead to progression of human pancreaticcancer (4). Hence, we postulated that functional interactionof Pim-3 and TCTP can promote human pancreatic carci-nogenesis. When HEK293T cells with endogenous TCTPproteinwere transiently transfectedwith the Pim-3 expressionvector, TCTP protein expression did not change (Fig. 2A).Moreover, TCTPprotein levels weremarginally affected evenwhen the amounts of Pim-3 (Fig. 2B) or incubation periodswere increased (data not shown). Furthermore, when Mia-Paca-2 cells with a low level of Pim-3 protein were transiently

293TA B

C D

E F

293T

Pim-3

Pim-3

TCTP

β-Actin

Pim-3

Pim-3

TCTP

β-Actin

β-Actin

Pim-3

TCTP

Pim-3 shRNA

β-Actin

Pim-3 0 0.1 0.2 0.3 0.4 0.5 μg

Pim-3

TCTP

β-Actin

− +

− + − +MiaPaca-2

MiaPaca-2

Stable Pim-3

TCTP

Phospho-TCTP(ser46)

Pim-3

P-Bad112

Bad

β-Actin

Stable Pim-3 shRNA

TCTP

Phospho-TCTP(ser46)

Pim-3

P-Bad112

Bad

PCI55

− +PCI55

− +

Figure 2. Pim-3 has no effect onTCTP expression orphosphorylation. A, B, cell lysateswere obtained from HEK293T cellstransiently transfected withpcDNA4-Pim-3 expression vector,empty pcDNA4 vector (A), or theindicated amounts of Pim-3expression vector (B). The resultantlysates were subjected toimmunoblotting with the indicatedantibodies. C, D, cell lysates wereobtained from MiaPaca-2 cellstransiently transfected withpcDNA4-Pim-3 expression orempty pcDNA4 vector (C) or fromPCI55 cells transiently transfectedwith Pim-3 shRNA or scrambledshRNA (D). The resultant lysateswere subjected to immunoblottingwith the indicated antibodies. E, F,cell lysates were obtained fromMiaPaca-2 cells stablyoverexpressing pMEI-5-Pim-3 orempty pMEI-5 vector (E) or fromPCI55 cells stably transfected withPim-3 or scramble shRNA vectors(F). The resultant lysates weresubjected to immunoblotting withthe indicated antibodies.

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1512

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

transfected with Pim-3, TCTP expression was not enhanced(Fig. 2C). When PCI55 cells with a high level of Pim-3protein was depleted of endogenous Pim-3, TCTP expressiondid not change discernibly (Fig. 2D). It was previouslyreported that TCTP can be phosphorylated mainly atSer46, but not at Thr65 (26). Pim-3 overexpression andablation of Pim-3 expression increased and decreased theamount of phospho-Ser112 Bad, respectively, but with feweffects on the levels of phospho-TCTPser46 or total amount ofBad in human pancreatic cancer cells (Fig. 2E and F). Theseobservations indicate that Pim-3 hasminimal effect onTCTPexpression or phosphorylation states.

TCTP enhances the protein stability of Pim-3 byblocking ubiquitin–proteasome degradation of Pim-3As TCTP can stabilize Mcl-1 protein (15), we assumed

that TCTP may enhance Pim-3 protein levels by stabi-lizing Pim-3. To address this possibility, we transientlycotransfected HEK293T cells with a fixed amount of Pim-3 and increasing quantities of the His-tagged TCTPexpression vectors. Pim-3 protein levels increased as theamounts of the transfected His-tagged TCTP vectors wereincreased (Fig. 3A). Moreover, ablation of endogenousTCTP protein significantly decreased the amount of Pim-3 protein but not Akt, indicating the specificity of theTCTP-mediated effect on Pim-3 levels (Fig. 3B). Theelevation of these protein levels could be either because ofenhanced mRNA expression, protein synthesis, or inhi-bition of protein degradation. We found that TCTPsiRNA treatment did not affect Pim-3 mRNA levels (Fig.3C). On the contrary, transient transfection of the TCTPexpression vector in HEK293T cells delayed the degra-dation of Pim-3 protein (Fig. 3D). Conversely, ablation ofTCTP with siRNA enhanced the degradation of Pim-3protein, compared with scrambled siRNA in PCI55 cells(Fig. 3E). Moreover, TCTP siRNA-mediated enhance-ment in Pim-3 protein degradation was abrogated by aproteasomal inhibitor, MG132 (Fig. 3F). Furthermore,TCTP siRNA treatment promoted Pim-3 ubiquitinationin PCI55 cells (Fig. 3G). Taken together, these resultsindicate that TCTP can increase Pim-3 protein stability byblocking the ubiquitin-proteasome-mediated degradationof Pim-3 in human pancreatic cancer cells.

Ablation of TCTP protein prevents Pim-3–mediatedtumor growth by arresting cell-cycle progression andenhancing apoptosisAblation of TCTP can destabilize Pim-3 protein in human

pancreatic cancer cells, whereas reduction of Pim-3 expres-sion can arrest cell-cycle progression and promote apoptosis(4). Hence, we investigated whether ablation of endogenousTCTP would affect the cell cycle and apoptosis in humanpancreatic cancer cells. TCTP shRNA, but not scrambledshRNA treatment markedly diminished Pim-3 proteins inPCI55, MiaPaca-2, and SW1990 cell lines (Fig. 4A) andsignificantly retarded cell proliferation compared with scram-ble shRNA (Fig. 4B). TCTP shRNA treatment increased theratio of G0–G1 phase cell population with reduced S and

G2–M phase population, compared with scrambled shRNAtreatment (Fig. 4C). Moreover, the stable transfection ofTCTP shRNA resulted in a markedly higher ratio of apo-ptotic cells as evidenced by enhanced phosphatidylserineexternalization (Fig. 4D–F). Furthermore, in comparison toa subcutaneous injection with control shRNA-transfectedcells, subcutaneous injectionwith TCTP shRNA-transfectedcells resulted in a lower tumor formation frequency and asmaller tumor mass in nude mice 4 weeks after injection(43% for MiaPaca-2-TCTP shRNA cells and 67% forSW1990-TCTP shRNA cells, respectively; 100% for bothof MiaPaca-2-scramble shRNA and SW1990-scrambleshRNA cells; Fig. 4G andH). Consistent with these observa-tions, ablation of TCTP decreased Pim-3 expression and theamounts of phosphorylated p21, the molecules that partic-ipate in cell progression at G0–G1 phase at the tumor sites inthe injected mice (ref. 27; Fig. 4I). Cyclin B1 and Cdc25C,molecules that participate in cell progression at G2–Mphase,were also decreased (Fig. 4I). Similarly, ablation of TCTPdecreased Pim-3 expression, and eventually diminished theamounts of phosphor-Ser112-Bad and Bcl-XL at the tumorsites in the injectedmice, without any effects on the amountsof Bad (Fig. 4I) and someother proteins relatedwith cell cycle(data not show).Collectively, these observations indicate thatTCTP can stabilize Pim-3 and promote cell-cycle progres-sion and cell viability, eventually promoting human pancre-atic carcinogenesis.

Clinical relevance of TCTP and Pim-3 expression inpancreatic ductal adenocarcinomaFinally, to investigate the clinical relevance of TCTP and

Pim-3 expression, we performed immunohistochemistry on148 resected pancreatic ductal adenocarcinoma (PDAC)tissues. TCTP and Pim-3 staining patterns were cyto-plasmic, and bothTCTP and Pim-3 expressionwere positivein more than 90% of PDAC tissues. Immunofluorescenceanalysis showed that TCTP and Pim-3 localized to thecytoplasm and colocalized in human pancreatic cancertissues (Fig. 5A). Moreover, TCTP protein was abundantlydetected in malignant ductal epithelium cells and somepancreatic acinar cells, but not in normal pancreatic ductepithelial cell adjacent to tumor foci (Fig. 5B). Positivereactions were not observed when control IgG was used asthe primary antibody instead of the anti-TCTP antibody(data not shown), indicating the specificity of the reaction.Furthermore, the expression patterns of TCTP and Pim-3exhibited perfect concordance in serial sections of humanpancreatic cancer tissues (Fig. 5C). TCTP expression sig-nificantly correlated with Pim-3 expression in 148 humanpancreatic ductal adenocarcinoma specimens (Spearmancorrelation coefficient ¼ 0.518; P < 0.01; Fig. 5D). Fur-thermore, higher TCTP and Pim-3 expression in PDACwassignificantly associated with an advanced tumor stage (P ¼0.001 and 0.008; Table 1). Finally, higher TCTP expressionalso correlated with nodal metastasis (P ¼ 0.024), butneither TCTPnor Pim-3 expression did show any significantcorrelation with age, gender, tumor location, tumor size, andtumor differentiation (Table 1). These observations suggest

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1513

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

Pim-3

PCI55

TCTP-His

A

C

D

E

F G

B0 0.5 1 1.5

exo.

endo.

μg

Pim-3

TCTP

TCTP

TCTP

β-Actin

β-Actin

β-Actin

TCTPPCI35 PCI55

Pim-3

TCTP

AKT

β-Actin

Scramble siRNA

Scramble siRNA

Scramble-siRNA

8

7

6

5

4

3

2

1

0

120

100

80

60

40

20

0

120

100

80

60

40

20

0

TCTP-siRNA

Scramble-siRNA

TCTP-siRNA

Scramble siRNA

TCTPmRNA

TCTP+

TCTP–

Pim-3mRNA

Rela

tive

mR

NA

leve

l

0 20 40 60 80 100 120 140 min

0 20 40 60 80 100 120 140 min

PCI55

170 kd

IP: Pim-3

WB: UB

TCTP siRNA − +

40 kd

Pim-3

Perc

enta

ge o

f P

im-3

pro

tein

leve

lP

erc

enta

ge o

f P

im-3

pro

tein

leve

l

Pim-3 Pim-3

+ + + + + + − − − − − −0 10 20 40 60 120 0 10 20 40 60 120

exo

endo

293T

Scramble siRNA

0 10 20 40 60 120 0 10 20 40 60 120

TCTP siRNA

PCI55

PCI55

TCTP-His

CHX (min)

Pim-3

TCTP

β-Actin

β-Actin

CHX (min)

Pim-3

TCTP

MG-132

TCTP siRNA

Pim-3

TCTP

− −+ −

+ ++ −

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1514

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

that enhanced TCTP and Pim-3 expressionmay be involvedin the malignant progression of human pancreatic cancer.

DiscussionWe previously observed that a proto-oncogene with ser-

ine/threonine kinase activity, Pim-3, is aberrantly expressedin various malignant lesions, but not normal tissues ofendoderm-derived organs such as the liver, pancreas, colon,and stomach (3–5, 6). Moreover, it can contribute totumorigenesis by inhibiting the apoptosis of tumor cellsand promoting their cell-cycle progression. Pim kinasefamily consists of 3 members, Pim-1, Pim-2, and Pim-3,which exhibit marked sequence similarity, especially in theirkinase domains. As Pim kinases do not possess a regulatorydomain and are constitutively active when they are expressed(28), the activity of Pim kinases is largely regulated attranscriptional, translation, and posttranslational levels(29). Furthermore, a very short half-life of their mRNAand protein (30) suggests the importance of regulation ofPim kinase protein levels.Pim-1 and Pim-3 have been shown to bind to the serine/

threonine protein phosphatase 2A (PP2A), resulting in theirdephosphorylation, ubiquitination, and proteasomal degra-dation (31, 32). Moreover, heat shock 70-kDa protein 1A/B(HsP70) binds ubiquitylated Pim-1 and promotes its pro-teasomal degradation. On the contrary, heat shock protein90b (HsP90) can stabilize Pim-1 protein by binding to it,and the inhibition of HsP90 induced rapid degradation ofPim-1 (33, 34). However, the molecular mechanisms ofPim-3 expression in carcinogenesis still remain largelyunknown. To define the detailed regulatory machinery, weconducted yeast 2-hybrid screening to identify Pim-3–inter-acting proteins and demonstrated for the first time thatTCTP directly interacts with Pim-3. TCTP is a multifunc-tional protein and can interact with many cellular proteins.TCTP binds to p53 to promote its proteasomal degradation(24, 25), whereas TCTP interacts with Mcl-1 to enhance itsstability (15). Indeed, Pim-3 protein stability is enhanced byits interaction with TCTP, which can block the ubiquitin-proteasome-mediated degradation of Pim-3.Comparisons of TCTP sequences in 24 eukaryotes (35)

revealed the presence of 2 highly conserved regions in TCTP

protein; one region from 45 to 55 residues and anotherregion from 129 to 147 residues. Several proteins caninteract with TCTP by binding to either region (Supple-mentary Table S3; refs. 26 and 35–39). We proved that theN-terminal portion of TCTP (residues 1–69) and the C-terminal portion of Pim-3 were required for their physicalinteraction. Although the crystal structure of Pim-3 has notyet been determined, Pim-3 seems to exhibit a similarstructure to Pim-1, because of its extraordinarily high aminoacid sequence similarity with Pim-3. The Pim-1 kinasestructure adopts a 2-lobe kinase fold with a deep andintervening cleft (28). The N-terminal and C-terminal lobesare composed of b-sheets and a-helices, respectively, where-as the 2 domains are connected through the hinge region(residues 121–126). TCTP could interact with the C-ter-minal but not the N-terminal portion-possessing Pim-3mutant, indicating that the a-helices in the C-terminalportion are mainly involved in binding to TCTP.Pim kinases can phosphorylate p27Kip1 and regulate its

expression at both transcriptional and posttranslationallevels, to promote tumorigenesis (40). Moreover, Pim-2overexpression in HCT116 cells leads to enhanced phos-phorylation of p21 to increase its stability (41). Furthermore,Pim-3 can augment protein synthesis (13) and regulatetranscriptional activity of Myc (30). However, overexpres-sion or ablation of Pim-3 failed to induce any changes inTCTP protein expression or phospho-TCTPser46 levels.Thus, Pim-3 has few effects on TCTP expression or phos-phorylation levels, even though Pim-3 can bind to TCTP.TCTP was overexpressed in many human cancer tissues

including liver (42), colorectal (43), prostate (44), breast(45), and lung cancer (46). Similarly, TCTP was detected inpancreatic cancer cell lines, and malignant duct epithelialcells and some normal acinar cells but not normal pancreaticduct epithelial cells adjacent to tumor foci.Moreover, TCTPexpression positively correlated with Pim-3 in pancreaticcancer cell lines and tissues. Furthermore, patients with highTCTP and Pim-3 expression often had advanced tumorstage. Thus, TCTP may promote pancreatic cancer bypreserving Pim-3 expression.All Pim kinase members can bind to and phosphorylate

CDK inhibitor, p27, at its threonine residues and induce the

Figure 3. TCTP enhances the protein stability of Pim-3 by blocking the ubiquitin–proteasome degradation of Pim-3. A, pcDNA4-Pim-3 expression vector(0.5 mg) was cotransfected with the indicated amount of pcDNA4-TCTP-His or empty pcDNA4 vector into HEK293T cells. After 48 hours, cell lysates wereanalyzed by Western blotting with the indicated antibodies. B, the cell lysates were obtained from PCI35 or PCI55 cells transiently transfected with TCTP orscramble siRNA. The resultant lysates were subjected to immunoblotting with the indicated antibodies. C, immunoblotting for TCTP and real-time qRT-PCRfor TCTPandPim-3 in TCTPsiRNA-treatedPCI55 cells (left). Cell lysateswere obtained fromPCI55 cells transiently transfectedwith TCTPor scramble siRNA.After 48 hours, the cell lysateswereobtained andsubjected to immunoblotting. Real-timeRT-PCRanalysiswasperformed toquantify TCTPandPim-3mRNAlevels as described in Materials and Methods. The results are expressed as mean � SD. ��P < 0.01 versus scrambled siRNA (right). D, pcDNA4-Pim-3expression vector (0.5 mg) was cotransfected with pcDNA4-TCTP-His (þ; 1.5 mg) or empty pcDNA4 (�; 1.5 mg) vector into HEK293T cells. At 24 hoursafter transfection, the cells were treatedwith cycloheximide (CHX, 30mg/mL) for various time intervals. Cell lysateswere then analyzedby immunoblottingwithindicated antibodies (left). ThepercentageofPim-3protein levelwasdeterminedusingdensitometry scanning (NIH ImageJsoftware) (right). E, TCTPor controlscramble siRNA was transfected transiently into PCI55 cells. At 24 hours after transfection, the cells were treated with cycloheximide for various times asindicated. Cell lysates were then analyzed by immunoblotting with the indicated antibodies (left). The percentage of Pim-3 protein level was determined usingdensitometry scanning (NIH ImageJ software) (right). F, PCI55 cells were transiently transfected with TCTP (þ) or scrambled siRNA (�). At 24 hoursafter transfection, the cells were treated with either dimethyl sulfoxide or MG132 (1 mg/mL) for an additional 4 hours. Cell lysates were obtained and subjectedto the immunoblottingwith the indicated antibodies.G, PCI55 cells were transiently transfectedwith TCTPsiRNA (þ) or scrambled siRNA (�). At 24 hours afterthe transfection, the cell lysates were prepared and immunoprecipitated with anti-Pim-3 antibodies. The immune complex was then analyzed byimmunoblotting with anti-ubiquitin (UB) or Pim-3 antibodies.

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1515

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

PCI55

A B

C

E

G I

H

F

D

MiaPaca-2 SW1990

TCTP shRNA

TCTP

Pim-3

β-Actin

scramble shRNA scramble shRNA scramble

Scramble shRNA9876543210

20

15

10

5

0

25

20

15

10

5

0

TCTP shRNA

Scramble shRNA

TCTP shRNAScramble shRNA

TCTP shRNA

Time after passage (d)0 1 2 3 4 5

Time after passage (d)0 1 2 3 4 5

Time after passage (h)

PCI55 MiaPaca-2 SW1990

0 12 24 36 48 60 72

PCI55 MiaPaca-2 SW1990

8.5

50.5

36.6

4.411

44.2

43.3

1.511.5

24.7

58.6

5.211

59.3

21.6

8.115.4

33.7

48.9

2 5

56.6

26

12.4100%

80%

Pe

rce

nta

ge

of

ce

lls

60%

40%

20%

0%

Scram

ble-

shRNA

TCTP-s

hRNA

Scram

ble-

shRNA

TCTP-s

hRNA

Scram

ble-

shRNA

TCTP-s

hRNA

G2/M

S

G0/G1

subG1

PCI55-scramble-shRNA PCI55-TCTP-shRNA

SW1990-scramble-shRNA

Pro

pid

ium

iodid

e

SW1990-TCTP-shRNA

Pro

pid

ium

iodid

e

Cell

via

bili

ty r

atio

Cell

via

bili

ty r

atio

Cell

via

bili

ty r

atio

Annexin V Scramble shRNA

TCTP shRNA

TCTP shRNA

Scramble shRNA

TCTP

TCTP

Pim-3

Cdc25C

Cyclin B1

P-p21T145

p21

P-Bad112

Bad

Bcl-XL

β-Actin

shRNA Scramble shRNA

shRNA

Scramble

Scramble

Perc

enta

ge o

f apopto

sis

shRNATCTP

shRNA Scramble

16

14

12

10

8

6

4

2

0

8

7

6

5

4

3

2

1

0

Perc

enta

ge o

f apopto

sis

Perc

enta

ge o

f apopto

sis

Annexin V

PCI55 MiaPaca-2 SW1990

14

12

10

8

6

4

2

0

0 5 10 15 20 25 30Days

0 5 10 15 20 25 30Days

250

200

150

100

50

0

180

160

140

120

100

80

60

40

20

0

Scramble-shRNA

MiaPaca-2 SW1990

TCTP-shRNA

N = 7

N = 3Scramble-shRNA

Tu

mo

r vo

lum

e (

mm

3)

Tu

mo

r vo

lum

e (

mm

3)

TCTP-shRNA

N = 9

N = 6

MiaPaca-2-scramble-shRNA MiaPaca-2-TCTP-shRNA

Annexin V

Pro

pid

ium

iodid

e

MiaPaca-2

Scramble shRNA TCTP shRNA Scramble shRNA TCTP shRNA

SW1190

# Mice

Tumorigenic

mice

Tumorigenic

ability

7 7 9 9

7 3 9 6

100% 43% 100% 67%

Figure 4. The effects of ablation of endogenous TCTP on cell viability and cell apoptosis of human pancreatic cancer cell lines. A, TCTP or scramble shRNAwas stably infected into PCI55, MiaPaca-2, or SW1990 cells using a lentivirus vector. TCTP and Pim-3 expression was detected by immunoblotting. b-Actinwas used as an internal control. B, PCI55, MiaPaca-2, or SW1990 cells were stably transfected with TCTP shRNA or scramble shRNA. Three thousand of theresultant cells were plated in a 96-well microplate. After 12 hours, when the cells were adhered to the microplate, was designated as time 0. The cellnumbers were determined as the indicated time intervals using the Cell Counting Kit-8 reagent and the ratios were compared with time 0. Point, mean (n ¼ 5);bars, SD. �, P < 0.05 versus Scramble shRNA. C, PCI55, MiaPaca-2, or SW1990 cells were stably transfected with TCTP shRNA or scrambled shRNA. Theproportion of cells in each cell-cycle phase was determined as described in Materials and Methods. (Legend continued on the following page.)

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1516

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

binding of p27 to 14-3-3 protein, resulting in its nuclearexport and proteasome-dependent degradation (40). More-over, Pim-1 can promote cell-cycle progression by phos-

phorylating and modulating the functions of moleculesinvolved in cell-cycle progression such as Cdc25A,cyclinD1-associated kinases (47), Cdc25C-associated kinase

TCTPA

B

C

D

Pim-3 DAPI

Merged MergedTCTP/Pim-3 TCTP/Pim-3/DAPI

Normal pancreatic duct epithelium

Adjacent to tumor foci Pancreatic adenocarcinoma

TCTP1 2 3 4 5

50 μmol/L

50 μmol/L 50 μmol/L

50 μmol/L 50 μmol/L 50 μmol/L

50 μmol/L 50 μmol/L

50 μmol/L 50 μmol/L 50 μmol/L 50 μmol/L 50 μmol/L

50 μmol/L 50 μmol/L 50 μmol/L 50 μmol/L 50 μmol/L

Pim-3

TCTP expression positive correlated with Pim-3 expression

Pim-3

TCTP

60

40

20

−++++++

− + ++ +++

Co

un

t (c

ase

)

Pim-3 expression

Total

TCTP expression

Total 14 59 65 10 148

−

− +

+

+ +

+ +

+ + +

+ + +

2

7

3

2 4 12 8 26

19 45 2 69

36 6 0 49

0 2 0 4

Figure 5. Clinicopathologicalanalysis of TCTP and Pim-3expression in PDAC. A,colocalization of TCTP and Pim-3in human pancreatic cancer tissue.Human pancreatic cancer tissueswere immunostained with acombination of anti-TCTPandanti-Pim-3 antibodies as described inMaterials and Methods. Thefluorescent images were digitallymerged. B, TCTP expression wasdetected on the section of normalpancreatic duct epitheliumadjacent to tumor foci (left andmiddle panels) and pancreaticadenocarcinoma (right panel)by immunohistochemistry.C, TCTP and Pim-3expression was analyzed byimmunohistochemistry on serialsections of different humanpancreatic cancer tissues. D,cross-tabulation of TCTP andPim-3 expression detected byimmunohistochemistry in 148pancreatic ductal adenocarcinomatissues. Correlation between TCTPand Pim-3 was significant onSpearman correlation analysis(Spearman correlation coefficient¼ 0.518, P < 0.01).

(Continued.) D–F, apoptosis of PCI55, MiaPaca-2, and SW1990 cells, which were stably transfected with TCTP or scrambled shRNA. Cells were stained with acombination of propidium iodide and Annexin V as described in Materials and Methods. The number in each quadrant indicates the proportion of the cellspresent in thequadrant. �,P< 0.05, ��,P< 0.01 versusScramble shRNA.G, incidenceof tumor formation in nudemice inducedby injectionofMiaPaca-2orSW1990cells, which were stably transfected with TCTP or scramble shRNA. H, tumor sizes were measured twice a week. The mean and SD were calculated and areshownhere. �,P< 0.05 versusScramble shRNA. I, cell lysateswereobtained fromPCI55,MiaPaca-2,andSW1990cells,whichwere stably transfectedwith TCTPorscramble shRNA. The resultant lysates were subjected to immunoblotting with the indicated antibodies as described in Materials and Methods.

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1517

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

1 (C-TAK1; ref. 48), and G1-specific inhibitor p21 (Waf;ref. 49). Given the high sequence identity between Pim-1and Pim-3, it is likely that Pim-3 can also modulate thesemolecules like Pim-1. Indeed, cell-cycle progression isaccelerated in hepatocytes of transgenic mice, whichexpress human Pim-3 cDNA selectively in hepatocytes(7). Moreover, a small-molecule Pim-3 kinase inhibitormarkedly retarded in vitro growth of human pancreaticcancer cell lines by inducing G2–M arrest (50), suggestinga potential role for Pim-3 in cell-cycle progression. Con-sistently, ablation of TCTP decreased the amounts ofPim-3, Cdc25C, cyclin B1, and phospho-p21T145, butnot the total amounts of p21. Furthermore, TCTP abla-tion arrested cell-cycle progression at the G1 phase inhuman pancreatic cancer cells, similar to when the cellswere depleted of Pim-3.Diraison and colleagues recently reported that increased

TCTP expression reduced the sensitivity of pancreatic b cellsto apoptosis (51).We previously observed that Pim-3 shRNAtreatment in vitro enhanced the apoptosis of various types ofcancer cells (3–5). Similarly, TCTP shRNA destabilized andinduced the degradation of Pim-3, and promoted the apo-ptosis of human pancreatic cancer cells. Moreover, TCTPshRNA-treated human pancreatic cancer cells exhibited a

weaker tumorigenic ability than scrambled shRNA-treatedcells, when they were injected into nude mice. Furthermore,we detected a much lower level of Pim-1 and Pim-2 mRNAthan Pim-3 mRNA in human pancreatic cancer cells that weexamined (Supplementary Fig. S1). Thus, it is likely thatTCTP can regulate the cell cycle and cell apoptosis mainly byPim-3–mediated in human pancreatic cancer cells.Several independent groups developed small-molecule

inhibitors against Pim kinases, including flavonol quercetar-getin, imidazole[1,2-b]pyridazines, bezylindene-thiazolidine-2,4-dione, 3,5-disubstituted indole derivatives, pyrazolo[3,4-g]quinoxaline derivatives, 1,6-dihydropyrazolo[4,3-c]carbazoles and 3,6-dihydropyrazolo[3,4-c]carbazoles der-ivatives (52), and pyrrolo[2,3-a]carbazole and pyrro-lo[2,3-g]indazoles derivatives (53, 54). Among them, 1,6-dihydropyrazolo[4,3-c]carbazoles, 3,6-dihydropyrazolo[3,4-c]carbazoles, and pyrrolo[2,3-g]indazoles could be used asan interesting molecular tool to study Pim-3 biologicalfunctions (52). Consistently, we also demonstrated thatstemonamide synthetic intermediates derivative can inhib-it Pim-3 as well as Pim-1 and Pim-2 activities and canreduce tumor growth in vivo xenograft models using ahuman pancreatic cancer cell line without causing majoradverse effects (55, 56). TCTP but not Pim-3 expression

Table 1. Correlation of TCTP and Pim-3 expression with clinicopathological features in 148 PDACspecimens

TCTP expression Pim-3 expression

Clinicopathological variable Patients �/þ þþ/þþþ P �/þ þþ/þþþ P

Age�62 y 73 31 42 0.096 35 38 0.741>62 y 75 22 53 38 37

GenderFemale 68 24 44 0.904 32 36 0.611Male 80 29 51 41 39

Tumor locationHead 76 32 44 0.101 38 38 0.866Body/tail 72 21 51 35 37

Tumor size�2 cm 34 10 24 0.516 16 18 0.5032–4 cm 64 26 38 35 29>4 cm 50 17 33 22 28

Nodal metastasis0 88 38 50 0.024a 45 43 0.5931 60 15 45 28 32

Tumor differentiationWell/moderate 100 34 66 0.507 47 53 0.414Poor 48 19 29 26 22

TNM stageI/II 118 49 65 0.001a 63 51 0.008a

III/IV 30 4 30 10 24

Abbreviation: TNM, tumor node metastasis.aP < 0.05.

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1518

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

was well correlated with nodal metastasis in humanpancreatic cancer patients. Thus, given the fact that TCTPcan bind and modulate several molecules, TCTP cancontribute to pancreatic cancer development and progres-sion besides its effects on Pim-3. If so, the targeting ofTCTP may supplement Pim-3 inhibitors for the treat-ment of various types of cancer, which exhibit enhancedPim-3 and TCTP expression.

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

Authors' ContributionsConception and design: Q.-X. Ni, N. Mukaida, Y.-Y. LiAcquisition of data (provided animals, acquired and managed pati-ents, provided facilities, etc.): F. Zhang, B. Liu, Z. Wang, X.-J. Yu,Q.-X. NiAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): F. Zhang, W.-T. Yang

Writing, review, and/or revision of the manuscript: Q.-X. Ni, W.-T. Yang,N. Mukaida, Y.-Y. LiAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): B. LiuStudy supervision: Q.-X. Ni, Y.-Y. Li

AcknowledgmentsThe authors thank Dr. Y. Qin, H.-Y. Gu, P. Zhang, Y.-H. Xin, and Y. Cao of the

Cancer Research Institute, Fudan University Shanghai Cancer Center for theirtechnical support.

Grant SupportThis work was supported in part by the National Science Foundation of China

(NSFC) (30973476, 812727), the Shanghai Pujiang Program (KW201028464),Fudan University "985 Project" Phase III Cancer Research Projects II (985III-YFX0102), and Shanghai Committee of Science and Technology (12DZ2260100).

The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received July 23, 2013; revised September 23, 2013; accepted October 10, 2013;published OnlineFirst October 28, 2013.

References1. Konietzko U, Kauselmann G, Scafidi J, Staubli U, Mikkers H, Berns

A, et al. Pim kinase expression is induced by LTP stimulation andrequired for the consolidation of enduring LTP. EMBO J 1999;18:3359–69.

2. Deneen B, Welford SM, Ho T, Hernandez F, Kurland I, Denny CT.PIM3 proto-oncogene kinase is a common transcriptional tar-get of divergent EWS/ETS oncoproteins. Mol Cell Biol 2003;23:3897–908.

3. Fujii C, Nakamoto Y, Lu P, Tsuneyama K, Popivanova BK, KanekoS, et al. Aberrant expression of serine/threonine kinase Pim-3in hepatocellular carcinoma development and its role in the pro-liferation of human hepatoma cell lines. Int J Cancer 2005;114:209–18.

4. Li YY, PopivanovaBK,Nagai Y, IshikuraH, Fujii C,MukaidaN. Pim-3, aproto-oncogene with serine/threonine kinase activity, is aberrantlyexpressed in human pancreatic cancer and phosphorylates bad toblock bad-mediated apoptosis in human pancreatic cancer cell lines.Cancer Res 2006;66:6741–7.

5. Popivanova BK, Li YY, Zheng H, Omura K, Fujii C, Tsuneyama K, et al.Proto-oncogene, Pim-3 with serine/threonine kinase activity, is aber-rantly expressed in human colon cancer cells and can prevent Bad-mediated apoptosis. Cancer Sci 2007;98:321–8.

6. Zheng HC, Tsuneyama K, Takahashi H, Miwa S, Sugiyama T, Popi-vanova BK, et al. Aberrant Pim-3 expression is involved in gastricadenoma-adenocarcinoma sequence and cancer progression. J Can-cer Res Clin Oncol 2008;134:481–8.

7. Wu Y, Wang YY, Nakamoto Y, Li YY, Baba T, Kaneko S, et al.Accelerated hepatocellular carcinoma development in mice expres-sing the Pim-3 transgene selectively in the liver. Oncogene 2010;29:2228–37.

8. Zha J, Harada H, Yang E, Jockel J, Korsmeyer SJ. Serine phosphor-ylation of death agonist BAD in response to survival factor results inbinding to 14-3-3 not BCL-X(L). Cell 1996;87:619–28.

9. Yang E, Zha J, Jockel J, Boise LH, Thompson CB, Korsmeyer SJ. Bad,a heterodimeric partner for Bcl-XL and Bcl-2, displaces Bax andpromotes cell death. Cell 1995;80:285–91.

10. Mukaida N, Wang YY, Li YY. Roles of Pim-3, a novel survival kinase, intumorigenesis. Cancer Sci 2011;102:1437–42.

11. Xu D, Cobb MG, Gavilano L, Witherspoon SM, Williams D, White CD,et al. Inhibition of oncogenic Pim-3 kinase modulates transformedgrowth and chemosensitizes pancreatic cancer cells to gemcitabine.Cancer Biol Ther 2013;14:492–501.

12. Wang C, Li HY, Liu B, Huang S, Wu L, Li YY. Pim-3 promotes thegrowth of human pancreatic cancer in the orthotopic nude mouse

model through vascular endothelium growth factor. J Surg Res 2013Jun 28. [Epub ahead of print]

13. Beharry Z, Mahajan S, Zemskova M, Lin YW, Tholanikunnel BG, Xia Z,et al. The Pim protein kinases regulate energy metabolism and cellgrowth. Proc Natl Acad Sci U S A 2011;108:528–33.

14. Li F, Zhang D, Fujise K. Characterization of fortilin, a novel antiapop-totic protein. J Biol Chem 2001;276:47542–9.

15. Liu H, Peng HW, Cheng YS, Yuan HS, Yang-Yen HF. Stabilization andenhancement of the antiapoptotic activity of mcl-1 by TCTP. Mol CellBiol 2005;25:3117–26.

16. Bohm H, Benndorf R, Gaestel M, Gross B, Nurnberg P, Kraft R, et al.The growth-related protein P23 of the Ehrlich ascites tumor: transla-tional control, cloning and primary structure. Biochem Int 1989;19:277–86.

17. Gross B, Gaestel M, Bohm H, Bielka H. cDNA sequence coding for atranslationally controlled human tumor protein. Nucleic Acids Res1989;17:8367.

18. Chitpatima ST, Makrides S, Bandyopadhyay R, Brawerman G. Nucle-otide sequence of a major messenger RNA for a 21 kDa polypeptidethat is under translational control in mouse tumor cells. Nucleic AcidsRes 1988;16:2350.

19. Thomas G, Thomas G, Luther H. Transcriptional and translationalcontrol of cytoplasmic proteins after serum stimulation of quiescentSwiss 3T3 cells. Proc Natl Acad Sci U S A 1981;78:5712–6.

20. Oikawa K, Ohbayashi T, Mimura J, Fujii-Kuriyama Y, Teshima S,Rokutan K, et al. Dioxin stimulates synthesis and secretion of IgE-dependent histamine-releasing factor. Biochem Biophys Res Com-mun 2002;290:984–7.

21. Nielsen HV, Johnsen AH, Sanchez JC, Hochstrasser DF, Schiotz PO.Identification of a basophil leukocyte interleukin-3-regulated proteinthat is identical to IgE-dependent histamine-releasing factor. Allergy1998;53:642–52.

22. Teshima S, Rokutan K, Nikawa T, Kishi K. Macrophage colony-stim-ulating factor stimulates synthesis and secretion of a mouse homologof a human IgE-dependent histamine-releasing factor by macro-phages in vitro and in vivo. J Immunol 1998;161:6356–66.

23. KoziolMJ,Gurdon JB. TCTP indevelopment andcancer. BiochemResInt 2012;2012:105203.

24. Rho SB, Lee JH, Park MS, Byun HJ, Kang S, Seo SS, et al. Anti-apoptotic protein TCTP controls the stability of the tumor suppressorp53. FEBS Lett 2011;585:29–35.

25. Amson R, Pece S, Lespagnol A, Vyas R, Mazzarol G, Tosoni D, et al.Reciprocal repression between P53 and TCTP. Nat Med 2012;18:91–9.

Stabilization of Pim-3 by TCTP

www.aacrjournals.org Mol Cancer Res; 11(12) December 2013 1519

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

26. Yarm FR. Plk phosphorylation regulates the microtubule-stabilizingprotein TCTP. Mol Cell Biol 2002;22:6209–21.

27. Zhang Y,Wang Z,Magnuson NS. Pim-1 kinase-dependent phosphor-ylation of p21Cip1/WAF1 regulates its stability and cellular localizationin H1299 cells. Mol Cancer Res 2007;5:909–22.

28. Qian KC, Wang L, Hickey ER, Studts J, Barringer K, Peng C, et al.Structural basis of constitutive activity and a unique nucleotide bindingmode of human Pim-1 kinase. J Biol Chem 2005;280:6130–7.

29. Amaravadi R, Thompson CB. The survival kinases Akt and Pim aspotential pharmacological targets. J Clin Invest 2005;115:2618–24.

30. Nawijn MC, Alendar A, Berns A. For better or for worse: the role of Pimoncogenes in tumorigenesis. Nat Rev Cancer 2011;11:23–34.

31. Ma J, Arnold HK, Lilly MB, Sears RC, Kraft AS. Negative regulation ofPim-1protein kinase levels by theB56beta subunit of PP2A.Oncogene2007;26:5145–53.

32. Losman JA, Chen XP, Vuong BQ, Fay S, Rothman PB. Protein phos-phatase 2A regulates the stability of Pim protein kinases. J Biol Chem2003;278:4800–5.

33. Shay KP, Wang Z, Xing PX, McKenzie IF, Magnuson NS. Pim-1 kinasestability is regulated by heat shock proteins and the ubiquitin-protea-some pathway. Mol Cancer Res 2005;3:170–81.

34. Mizuno K, Shirogane T, Shinohara A, Iwamatsu A, Hibi M, Hirano T.Regulation of Pim-1 by Hsp90. Biochem Biophys Res Commun2001;281:663–9.

35. Thaw P, Baxter NJ, Hounslow AM, Price C, Waltho JP, Craven CJ.Structure of TCTP reveals unexpected relationship with guaninenucleotide-free chaperones. Nat Struct Biol 2001;8:701–4.

36. YangY,YangF, XiongZ,YanY,WangX,NishinoM, et al. AnN-terminalregion of translationally controlled tumor protein is required for itsantiapoptotic activity. Oncogene 2005;24:4778–88.

37. CansC, Passer BJ, Shalak V,Nancy-Portebois V,Crible V, AmzallagN,et al. Translationally controlled tumor protein acts as a guanine nucle-otide dissociation inhibitor on the translation elongation factor eEF1A.Proc Natl Acad Sci U S A 2003;100:13892–7.

38. Yoon T, Jung J, Kim M, Lee KM, Choi EC, Lee K. Identification of theself-interaction of rat TCTP/IgE-dependent histamine-releasing factorusing yeast two-hybrid system. Arch Biochem Biophys 2000;384:379–82.

39. KimM, Jung Y, Lee K, KimC. Identification of the calcium binding sitesin translationally controlled tumor protein. Arch Pharm Res 2000;23:633–6.

40. Morishita D, Katayama R, Sekimizu K, Tsuruo T, Fujita N. Pim kinasespromote cell cycle progression by phosphorylating and down-regu-lating p27Kip1 at the transcriptional and posttranscriptional levels.Cancer Res 2008;68:5076–85.

41. Wang Z, Zhang Y, Gu JJ, Davitt C, Reeves R, Magnuson NS. Pim-2phosphorylation of p21(Cip1/WAF1) enhances its stability and inhibitscell proliferation in HCT116 cells. Int J Biochem Cell Biol 2010;42:1030–8.

42. ChanTH,Chen L, LiuM,Hu L, ZhengBJ, PoonVK, et al. Translationallycontrolled tumor protein induces mitotic defects and chromosomemissegregation in hepatocellular carcinomadevelopment. Hepatology2012;55:491–505.

43. Slaby O, Sobkova K, Svoboda M, Garajova I, Fabian P, Hrstka R, et al.Significant overexpression of Hsp110 gene during colorectal cancerprogression. Oncol Rep 2009;21:1235–41.

44. Arcuri F, Papa S, Carducci A, Romagnoli R, Liberatori S, Riparbelli MG,et al. Translationally controlled tumor protein (TCTP) in the humanprostate and prostate cancer cells: expression, distribution, and cal-cium binding activity. Prostate 2004;60:130–40.

45. Deng SS, Xing TY, Zhou HY, Xiong RH, Lu YG, Wen B, et al.Comparative proteome analysis of breast cancer and adjacentnormal breast tissues in human. Genom Proteom Bioinformat2006;4:165–72.

46. Kim JE, Koo KH, Kim YH, Sohn J, Park YG. Identification of potentiallung cancer biomarkers using an in vitro carcinogenesis model. ExpMol Med 2008;40:709–20.

47. Mochizuki T, Kitanaka C, Noguchi K, Muramatsu T, Asai A, KuchinoY. Physical and functional interactions between Pim-1 kinase andCdc25A phosphatase. Implications for the Pim-1-mediated acti-vation of the c-Myc signaling pathway. J Biol Chem 1999;274:18659–66.

48. Bachmann M, Hennemann H, Xing PX, Hoffmann I, Moroy T. Theoncogenic serine/threonine kinase Pim-1 phosphorylates and inhibitsthe activity of Cdc25C-associated kinase 1 (C-TAK1): a novel role forPim-1 at the G2/M cell cycle checkpoint. J Biol Chem 2004;279:48319–28.

49. Wang Z, Bhattacharya N, Mixter PF, Wei W, Sedivy J, Magnuson NS.Phosphorylation of the cell cycle inhibitor p21Cip1/WAF1 by Pim-1kinase. Biochim Biophys Acta 2002;1593:45–55.

50. Wang YY, Taniguchi T, Baba T, Li YY, Ishibashi H, Mukaida N.Identification of a phenanthrene derivative as a potent anticancer drugwith Pim kinase inhibitory activity. Cancer Sci 2012;103:107–15.

51. Diraison F, Hayward K, Sanders KL, Brozzi F, Lajus S, Hancock J, et al.Translationally controlled tumour protein (TCTP) is a novel glucose-regulated protein that is important for survival of pancreatic beta cells.Diabetologia 2011;54:368–79.

52. Suchaud V, Gavara L, Saugues E, Nauton L, Thery V, Anizon F, et al.Identification of 1,6-dihydropyrazolo[4,3-c]carbazoles and 3,6-dihydropyrazolo[3,4-c]carbazoles asnewPimkinase inhibitors. BioorgMed Chem 2013;21:4102–11.

53. Akue-Gedu R, Letribot B, Saugues E, Debiton E, Anizon F, Moreau P.Kinase inhibitory potencies and in vitro antiproliferative activities of N-10 substituted pyrrolo[2,3-a]carbazole derivatives. Bioorg Med ChemLett 2012;22:3807–9.

54. Gavara L, Suchaud V, Nauton L, Thery V, Anizon F, Moreau P.Identification of pyrrolo[2,3-g]indazoles as new Pim kinase inhibitors.Bioorg Med Chem Lett 2013;23:2298–301.

55. Li YY, Wang YY, Taniguchi T, Kawakami T, Baba T, Ishibashi H, et al.Identification of stemonamide synthetic intermediates as a novelpotent anticancer drug with an apoptosis-inducing ability. Int J Cancer2010;127:474–84.

56. Wang Z, Li XM, Shang K, Zhang P, Wang CF, Xin YH, et al. T-18, astemonamide synthetic intermediate inhibits Pim kinase activity andinduces cell apoptosis, acting as a potent anticancer drug. Oncol Rep2013;29:1245–51.

Zhang et al.

Mol Cancer Res; 11(12) December 2013 Molecular Cancer Research1520

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389

2013;11:1508-1520. Published OnlineFirst October 28, 2013.Mol Cancer Res   Fei Zhang, Bin Liu, Zhen Wang, et al.   Involvement in Pancreatic Cancer ProgressionA Novel Regulatory Mechanism of Pim-3 Kinase Stability and Its

  Updated version

  10.1158/1541-7786.MCR-13-0389doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://mcr.aacrjournals.org/content/suppl/2013/10/28/1541-7786.MCR-13-0389.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://mcr.aacrjournals.org/content/11/12/1508.full#ref-list-1

This article cites 55 articles, 17 of which you can access for free at:

  Citing articles

  http://mcr.aacrjournals.org/content/11/12/1508.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

  .pubs@aacr.org

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://mcr.aacrjournals.org/content/11/12/1508To request permission to re-use all or part of this article, use this link

on February 11, 2018. © 2013 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst October 28, 2013; DOI: 10.1158/1541-7786.MCR-13-0389