p53 Mutation Directs AURKA Overexpression via miR-25 and ... · miR-25 and FBXW7 in Prostatic Small...

Signal Transduction

p53 Mutation Directs AURKA Overexpression viamiR-25 and FBXW7 in Prostatic Small CellNeuroendocrine CarcinomaZhenLi1,YinSun1, XufengChen1, Jill Squires1, BehdokhtNowroozizadeh1,ChaozhaoLiang2,and Jiaoti Huang1

Abstract

Prostatic small cell neuroendocrine carcinoma (SCNC) is arare but aggressive form of prostate cancer that is negative forandrogen receptor (AR) and not responsive to hormonaltherapy. The molecular etiology of this prostate cancer variantis not well understood; however, mutation of the p53 (TP53)tumor suppressor in prostate neuroendocrine cells inactivatesthe IL8–CXCR2–p53 pathway that normally inhibits cellularproliferation, leading to the development of SCNC. SCNC alsooverexpresses Aurora kinase A (AURKA) which is considered tobe a viable therapeutic target. Therefore, the relationship ofthese two molecular events was studied, and we show that p53

mutation leads to increased expression of miR-25 and down-regulation of the E3 ubiquitin ligase FBXW7, resulting inelevated levels of Aurora kinase A. This study demonstratesan intracellular pathway by which p53 mutation leads toAurora kinase A expression, which is critically important forthe rapid proliferation and aggressive behavior of prostaticSCNC.

Implications: The pathogenesis of prostatic SCNC involves a p53and Aurora Kinase A signaling mechanism, both potentiallytargetable pathways. Mol Cancer Res; 13(3); 584–91. �2014 AACR.

IntroductionProstate cancer is the leading cause of cancer-related death for

men in western countries. Understanding the molecular mechan-isms of prostate carcinogenesis and progression is the foundationand a challenge for the development of effective therapy. Patientswith low grade and early stage of prostate cancers can be cured bysurgery or radiotherapy. For those with advanced and metastaticprostate cancers that are not amenable for local therapies, hor-monal therapy targeting androgen receptor (AR) pathway hasbeen the treatment of choice for many decades. Unfortunately,this therapy is not curative, and the cancer invariably progresses tocastration-resistant state with few therapeutic options.

The majority of human prostate cancers are classified as ade-nocarcinoma with the bulk tumor cells showing luminal differ-entiation including the expressionofAR andPSA. Interestingly, all

adenocarcinomas of the prostate contain some neuroendocrine(NE) cells (1, 2). Unlike the bulk tumor cells, the scattered NEtumor cells are usually quiescent. In contrast, an importanthistologic variant prostate cancer called small cell neuroendocrinecarcinoma (SCNC) is composed of NE tumor cells that are highlyproliferative and aggressive. Although SCNC is occasionally diag-nosed in patients without any previous history of prostate cancer,it more commonly occurs as a recurrent tumor in patients witha history of adenocarcinoma who have received hormonal ther-apy. It has been suggested that the novel drugs abiraterone andenzalutamide (formerly known asMDV3100) that further inhibitAR signaling will induce even more cases of SCNC.

We recently demonstrated that the IL8/CXCR2/p53 signalingpathway keeps the NE cells in adenocarcinoma in a quiescentstate, and mutant p53 inactivates this pathway, leading to hyper-proliferation of NE cells and the development of SCNC (3).Meanwhile, previous study also found that Aurora kinase A wasoverexpressed in the majority of cases of SCNC, indicating apotential role of AuroraKinaseA in thedevelopment of SCNC(4).

In this study, we provide evidence showing that p53 mutationleads to elevated expression of Aurora kinase A through regulationof miR-25 and FBXW7 (F-box and WD repeat domain containing7, E3 ubiquitin protein ligase) thus revealing a potential molec-ular mechanism of p53 mutation in promoting the rapid prolif-eration and aggressive behavior of NE tumor cells in prostaticSCNC.

Materials and MethodsCell lines

Human prostate LNCaP Clone FGC, PC-3, and NCI-H660 cellswere from American Type Culture Collection (ATCC) and wereauthenticated utilizing short tandem repeat profiling. LNCaPClone FGC cells were cultured in ATCC-formulated RPMI-1640

1Department of Pathology and Urology, Jonsson ComprehensiveCancer Center and Broad Center of Regenerative Medicine and StemCell Research, UCLA David Geffen School of Medicine, Los Angeles,California. 2Department of Urology, The First Affiliated Hospital ofAnhui Medical University, Hefei, China.

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

Z. Li and Y. Sun contributed equally to this article.

Current address for Y. Sun: Department of Radiation Oncology, University ofRochester Medical Center School of Medicine and Dentistry, Rochester, NY.

Corresponding Authors: Jiaoti Huang, UCLA David Geffen School of Medicine,10833 LeConte Ave., 13-226CHS, LosAngeles, CA90095. Phone: 310-267-2264;Fax 310-794-4161; E-mail: [email protected]; or Yin Sun,[email protected]

doi: 10.1158/1541-7786.MCR-14-0277-T

�2014 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 13(3) March 2015584

on October 3, 2020. © 2015 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst December 15, 2014; DOI: 10.1158/1541-7786.MCR-14-0277-T

http://mcr.aacrjournals.org/

medium supplemented with 10% FBS, 2 mmol/L L-glutamine,100 U/mL penicillin, and 100 mg/mL streptomycin in a humid-ified atmosphere of 5% CO2 maintained at 37�C. PC-3 cellswere cultured in ATCC-formulated F-12Kmediumwith 10% FBS.NCI-H660 cells were cultured in RPMI-1640 medium with 0.005mg/mL insulin, 0.01 mg/mL transferrin, 30 nmol/L sodiumselenite, 10 nmol/L hydrocortisone, 10 nmol/L beta-estradiol,4mmol/L L-glutamine, and 5%FBS (HITESmedium). NE1.8 cellswere provided by Dr. Ming-Fong Lin (5), and were cultured inphenol red–free RPMI 1640 supplemented with 10% charcoal-stripped FBS.

Nucleic acidsSmall interference RNA for TP53 was purchased from IDT

as predesigned siRNA: sense strand 50rCrCrArCrCrArUrCrCr-ArCrUrArCrArArCrUrArCrArUrGT30, antisense strand 50rCrA-rCrArUrGrUrArGrUrUrGrUrArGrUrGrGrArUrGrGrUrGrGrU-rA30. Small interference RNA for FBXW7: sense strand 50-rGrGrArGrUrGrGrArCrCrArGrArGrArArArUrUrGrCrUrUG C-30,antisense strand 50-rGrCrArArGrCrArArUrUrUrCrUrCrUrGrGr-UrCrCrArCrUrCrCrArG-30. cDNA of firefly luciferase was clonedinto pCI-Neo vector followed by 30 untranslated region (UTR) ofFBXW7, which was joined by two separate PCR fragments (leftfragment: 50CTAGTCTAGAAGAGCAGAAAAGATGAATTT30 and50TTAGAGGCACAGATGGCTCA30, right fragment: 50TTGTCC-CAACCCTGTACTGTA30 and 50CATGAAAAAACACATTTTATTG-CACTTAAGTTATAAG30) after restriction digestion with EcoRIfollowed by ligation. Mutant 30UTR of FBXW7 was constructedby theQuickChangemethod to change the sequence of consensusmiR-25 seed sequence of 50UGCAAU30 at two locations into amutant sequence of 50GGAUCC30. Plasmid DNA encoding wild-type p53 (pCDNA3.1-p53wt) was described previously (6). Plas-mid DNA encoding R175H mutation of p53 (DNp53) wasgenerated by mutagenesis PCR. These constructs were verifiedthrough restriction digestion and sequencing analysis.

Lentivirusp53 (R175H) was subcloned into the EcoRI site of FUCRW

lentiviral vectors (7). This construct was verified through restric-tion digestion and sequencing analysis. The lentivirus was pre-pared and titered as described (8). LNCaP cells were spin infectedat 1,800 rpm for 45minutes at room temperature. All procedureswereperformedunderUniversity ofCalifornia, LosAngeles, safetyregulations for lentivirus usage.

AntibodiesAnti–Aurora A kinase antibody was from Cell Signaling

Technology, anti-FBXW7 antibody was from Bethyl Laboratories,anti-p53 antibody and anti–c-Myc were from Santa Cruz Biotech-nology, anti-MYCN antibody was from Abgent, anti-GAPDHantibody was from GeneTex, Inc.

Immunoblot assayCells were washed with PBS and lysed in RIPA buffer (50

mmol/L Tris, pH 7.4, 150mmol/L NaCl, 1% Triton X-100, 0.5%deoxycholate, 0.1%SDS) containing SigmaFAST Protease Inhib-itor Cocktail (Sigma-Aldrich) for 15 minutes at 4�C. Cell lysateswere centrifuged and supernatants were collected. Equivalentamounts of proteins as measured by Bradford assay wereresolved on SDS-PAGE gels and transferred to PVDF mem-

branes. The resulting blots were blocked in 5% nonfat dry milkin PBS for 30 minutes followed by incubation with primaryantibody in 5% BSA overnight. Appropriate horseradish perox-idase (HRP)–conjugated secondary antibodies and SupersignalWest Femto chemiluminescent substrate (Thermo Fisher Scien-tific) were used to visualize antigen–antibody complexes.

siRNA transfectionTransfections were performed with negative control, TP53, or

Fbxw7 siRNA (IDT) using the Xfect siRNA Transfection Reagent(Clontech), according to the manufacturer's protocol.

Quantitative RT-PCRTotal RNA ormiRNAwas extracted from cells using the RNeasy

Mini Kit (Qiagen) per the manufacturer's instructions. Conver-sion to cDNA was achieved through the PrimeScript RT MasterMix (Takara). Quantitative RT-PCR was carried out using SYBRPremix Ex Taq II (Takara), 0.4 mmol/L oligonucleotide primers,and 0.1 mg cDNA. All primer sets for quantitative RT-PCR wereillustrated in Supplementary Table S1. miRNA quantification wasperformed using themiRCURY LNAUniversal RTmicroRNA PCRStarter Kit (Exiqon). Relative fold change in mRNA levels wascalculated after normalization to b-actin using the comparative Ct

method (9).

IHCFor immunohistochemical analysis of p53 and Aurora A

kinase, tissue sections were deparaffinized with xylene and rehy-drated through graded ethanol. Endogenous peroxidase activitywas blocked with 3% hydrogen peroxide in methanol for 10minutes. Heat-induced antigen retrieval (HIER) was carried outfor all sections in 0.01 mol/L citrate buffer, pH 6.0, using avegetable steamer at 95�C for 25 minutes. Mouse monoclonalanti-p53 antibody, clone 1801 (EMD,OP09-100UG) was dilutedwith BSA to a concentration of 1:50 and applied to the sections.Incubation was for 45 minutes at room temperature followed byanti-mouse secondary antibody (MACH 2 Mouse HRP-Polymer;Biocare Medical; MHRP520L) incubation for 30minutes at roomtemperature. Rabbit monoclonal Aurora kinase A antibody(Abcam; 1800-1) was diluted with BSA to a concentration of1:50 and applied to the sections. Incubation was for 1 hour atroom temperature followed by anti-rabbit secondary antibody(Dakocytomation Envision System Labelled Polymer HRP antirabbit, Cat.# 4003) incubation for 30 minutes at room temper-ature. Diaminobenzidine was then applied for 10 minutes atroom temperature to visualize p53 and Aurora Kinase A. Sectionswere counterstained with hematoxylin, dehydrated through grad-ed alcohols, and coverslipped. Immunohistochemical semiquan-titation was performed using the Quick score (Q) method (10).Results are scored by multiplying the percentage of positivecells (P) by the intensity (I) (0 ¼ no staining, 1 ¼ weak staining,2¼moderate staining, 3¼ strong staining). Formula is defined asQ ¼ P � I; maximum ¼ 300.

Immunofluorescence double stainingSlides were deparaffinized with xylene and rehydrated through

graded ethanol. HIER was carried out in 0.01mol/L citrate buffer,pH 6.00, using a vegetable steamer at 95�C for 25 minutes.Sections were permeabilized for 10 minutes with 0.25% TritonX-100 and rinsed with PBS. Blocking was done with 2% BSA for

p53 Regulates Aurora Kinase A

www.aacrjournals.org Mol Cancer Res; 13(3) March 2015 585




30 minutes at room temperature. Primary antibody mixtures(Aurora Kinase A 1:100 BSA þ p53 1:25 BSA) were applied for1 hour at room temperature. Slides were rinsed with PBS, and thesecondary antibodymixture (goat anti–Mouse-Alexa Fluor 488þgoat anti–rabbit-Alexa Fluor 568, both1:500BSA)was applied for1 hour at room temperature. Slides were rinsed with PBS andcoverslipped using VECTASHIELD HardSet Mounting Mediumwith DAPI (Vector, H-1500).

Statistical analysisStatistical analyses were performed using the Student t test with

the Excel 2013 software. Error bars indicate SD calculated fromthree independent experiments.

ResultsP53 mutation leads to increased expression of miR-25 inprostate cancer cells

We previously demonstrated that the quiescent NE cells inprostatic adenocarcinoma contain wild-type p53, whereas therapidly proliferating NE tumor cells of SCNC often containmutated p53 (3). We proposed that p53 mutation may play acritical role in the development of aggressive behavior ofprostatic SCNC, but the detailed mechanisms were unclear.P53 can regulate miRNA expression in cancer cells (11). Inglioblastoma cells, for example, p53 has been reported torepress the expression of miR-25 and -32 (12). Thus, it is quiteinteresting whether there is also a relationship between p53expression/function and the expression of miRNAs, such asmiR-25 and/or -32, and the interaction then contributes to thebiologic behavior of prostatic SCNC. We therefore tested this

hypothesis in prostate cancer PC-3 and LNCaP cells. Ourprevious study has shown that LNCaP cells are typical prostateadenocarcinoma cells, and PC-3 cells are characteristic of SCNC(13). In addition, LNCaP cells express wild-type p53 protein,and PC-3 cells contain truncated p53 mutation, which leads tothe absence of p53 protein expression.

We expressed either wild-type p53 protein in PC-3 cells, ormutant p53 that is defective in DNA binding (R175H) in LNCaPcells.We found that expression ofwild-type p53 protein repressedmiR-25 expression in PC-3 cells, and expression of R175Hmutantp53 protein increased miR-25 level in LNCaP cells, both withstatistical significances (Fig. 1A).However, we noticed that expres-sion of wild-type p53 protein in PC-3 cells did not cause obviouschanges in the expression ofmiR-32 (data not shown). In prostateNE cancer cell lineNCI-H660 that containswild-type p53,we alsoobserved that knockdown of p53 by siRNA or expression of thedominant-negative p53mutant both resulted in enhancedmiR-25expression (Fig. 1B).

We further examined whether the changes of miR-25 level inthese cells were associated with potential regulation of cellcycles induced by changes of p53 status. We found that expres-sion of dominant-negative p53 (DNp53) did not cause obviouschanges of cell cycle distribution in LNCaP cells, although miR-25 level changes observed. In PC-3 cells, however, expression ofwild-type p53 (WTp53) did induce a change in G2–M transition24 hours later after transfection of mammalian expressiveWTp53 construct. Interestingly, we observed statistically signi-ficant reduction of miR-25 levels in cells at this time point(Supplementary Fig. S1).

Taken together, these results suggest that p53 can regulatemiR-25 expression in prostate cancer cells, whereas mutant p53

PC-3A

C

B

miR-25

MW(kD) MW(kD)55 P53

FBXW7

AURKA

GAPDH

DNP53wtP53

FBXW7

AURKA

GAPDH

100

70

40

35

55

100

70

40

35

0miR-25

Controlwt p53

Control

NCI H660

NCI-H660 NCI-H660

DNp53

siCtrl

sip53

siCtrl

sip53

DN-p53

Empty ve

ctor

DN-p53

Empty ve

ctor

1.2

1

0.8

0.6

0.4

0.2

0

4.54

3.53

2.52

1.51

0.50

1.8

1.4

1

0.6

0.2

Rat

io o

f miR

-25/

miR

-191

Rat

io o

f miR

-25/

miR

-191LNCaP

Figure 1.p53 regulates miR-25, FBXW7, andAurora kinase A. A and B, miR-25expression changes in response top53 expression and mutation status.Plasmids encoding wild-type p53(left) or mutant p53 (right) weretransfected together withGFP-expression plasmid into PC-3 orLNCaP cells for 48 hours. GFP-positive cells were sorted and RNAswere then isolated for quantificationof miR-25 using qRT-PCR. Twoindependent triplicate experimentswere performed, and results arepresented as mean � SD. The ratio ofthemiR-25 overmiR-191 as an internalcontrol was plotted (A). siRNA forp53, or lentiviral vector for dominant-negative p53 was introduced intoNCI H660 cells for 48 hours, andmiR-25 was quantitated as in A (B).Asterisks, significant differences(P < 0.05). C, immunoblot analysisresults showing the changes ofproteins in response to p53regulation for 48 hours in NCIH660 cells.

Li et al.

Mol Cancer Res; 13(3) March 2015 Molecular Cancer Research586




or loss of p53 functions can cause elevated expression of miR-25expression.

P53 mutation leads to decreased expression of FBXW7 andoverexpression of Aurora kinase A

miR-25 hasmany potential targets including FBXW7 andWwp2(WW domain containing E3 ubiquitin protein ligase 2) (14).Indeed, we observed that overexpression of miR-25 could lead toreduced levels of both FBXW7 andWWP2 in PC-3 cells, as well asin NE1.8 cells, a variant of LNCaP cells that resemble NE cells(Supplementary Fig. S2; ref. 5). FBXW7 encodes an E3 ubiquitinligase whose substrates include several positive cell cycle regula-tors, such as MYCN (15), MYC (16, 17), Cyclin E (18, 19), andAurora kinase A (20, 21). Of them, Aurora Kinase A is over-expressed in prostatic SCNC and may play important roles in thedevelopment of aggressive prostate tumor (4). We thus furthertested whether p53 mutation or loss of function could causechanges of Aurora Kinase A expression via regulation of miR-25and FBXW7. Indeed, as shown in Fig. 1C, we found that knockingdown p53 with siRNA in NCI-H660 cells resulted in increase ofAurora kinase A expression. And expression of DNp53 in NCI-H660 led to a change of Aurora kinase A expression similar to p53knockdown. In addition, we observed reduced expression ofFBXW7 in NCI-H660 cells with these manipulations of p53expression.

Next, we determined if change of FBXW7 expression wouldaffect the protein level of Aurora kinase A. In NCI-H660 and PC-3cells, we found transfections of Fbxw7 siRNA decreased theexpression of FBXW7 protein, and increased levels of Aurorakinase A. Similar results were also observed inNE1.8 cells (Fig. 2).In these cells, silencing of Fbxw7 also resulted in elevated proteinlevels of its targets C-MYC and MYCN. Thus, these results sug-gested that FBXW7 may also function as the ubiquitin E3ligase targeting Aurora kinase A.

miR-25 mediates mutant p53–induced expression of Aurora Akinase

We next determined the potential role of miR-25 in the regula-tions of Aurora kinase A upon loss of p53 function in theseprostate cancer cells. For this, we cotransfected p53 siRNA withmiR-25 inhibitor into LNCaP cells. Figure 3A and B showed thatmiR-25 inhibitor prevented p53 knockdown–induced downregu-lation of FBXW7 and upregulation of Aurora kinase A in theseLNCaP cells, indicating that loss of function of p53 may regulateAurora kinase A expression through a linear pathway involving

miR-25 and FBXW7. To further clarify the roles of miR-25 in thisprocess, we transfected miR-25 into NE1.8 cells. As expected, ourresults showed that overexpression of miR-25 reduced FBXW7level and increased Aurora kinase A protein expression (Fig. 3C).

To determine whether miR-25 can regulate FBXW7 expres-sion through its likely target in its 30UTR region, we generated areporter construct where coding sequence for firefly luciferasewas followed by a 30UTR sequence of FBXW7, or by a mutant30UTR sequence of FBXW7 where the target sequences of miR-25 were destroyed. As shown in Fig. 3D, exogenous expressionof miR-25 caused approximately 50% reduction of the lucif-erase activity with the wild-type 30UTR of FBXW7, whereas noobvious changes of luciferase activity was observed with themutant 30UTR in response to the miRNA expression. Thus, ourresults indicated that miR-25 could directly regulate FBXW7expression through the miRNA binding site in its 30UTR.

We also tested the potential roles of miR-25 in the biologicbehaviors of SCNC PC-3 cells. Our results showed that inhibitionof miR-25 in PC-3 cells attenuates cell proliferation and reducedinvasion capability (Supplementary Fig. S3).

Taken together, our results suggested a likely signaling pathwaythroughwhich p53mutation induces upregulation ofmiR-25 anddownregulation of FBXW7, eventually leading to the overexpres-sionofAurora kinaseA,whichmaypromote rapid proliferation ofthe NE tumor cells of SCNC.

Coexpression of nuclear p53 and Aurora kinase A in humanSCNC tissue

We further verified the coexpressionof p53andAuroraKinaseAin 12 cases of human prostatic SCNC and an equal number ofcases of prostate adenocarcinoma. Our IHC study showed that,eight of the 12 prostatic SCNC cases had positive nuclear p53staining, which usually results from mutation of p53 withincreased p53 protein stability. Nine of the 12 cases also showedoverexpression of Aurora kinase A, of them, 6 cases were positivefor both nuclear p53 and Aurora kinase A overexpression (Fig.4A). We also noticed that all 12 cases of adenocarcinoma werenegative for both p53 nuclear staining and Aurora kinase Aoverexpression. Furthermore,we found the coexpression of nucle-ar p53 and Aurora kinase A in these SCNC tissues (Fig. 4B). Theseresults supported our hypothesis that p53mutationmight lead tothe overexpression of Aurora Kinase A in prostatic SCNC.

We have shown previously that p53 mutation is commonin prostatic SCNC and rare in untreated adenocarcinoma (3).Similarly, Rubin's group (22) showed that Aurora kinase A is

PC-3 NE1.8

MW(kD)

FBXW7

AURKA

GAPDH

FBXW7

AURKA

GAPDH

MYC

MYCN

FBXW7

AURKA

GAPDH

MYC

MYCN

100

70

40

35

MW(kD)100

70

40

35

70

55

MW(kD)100

70

40

35

70

55

NCI-H660

siCtrl

siFbx

w7

siCtrl

siFbx

w7

siCtrl

siFbx

w7

Figure 2.Fbxw7 regulates Aurora kinase Aexpression. Control or FBXW7 siRNAat 100 nmol/L was transfected intoNE1.8, PC-3, or NCI-H660 cells. Aurorakinase A and FBXW7 were examinedvia immunoblot analysis. MYCN andMYC are included as known validatedtargets of FBXW7.






overexpressed in prostatic SCNC but not in adenocarcinoma.Therefore, we performed analysis for miR-25 expression betweenthe two types of tumors. In this study, total RNA was isolatedfrom the paraffin section of the cancerous area from 12 cases ofSCNC and an equal number of adenocarcinoma cases. Thelevels of miR-25 were measured and normalized to miR-191 asthe internal control, and the ratios of these two miRNAs werethen plotted. The box-and-whisker plots showed that almost halfof the cases of prostatic SCNC have higher levels of miR-25expression when compared with prostate adenocarcinoma(Fig. 4C). Protein overexpression of Aurora kinase A in humanprostatic SCNC was confirmed by IHC (Fig. 4D and E). Thus datafrom the human prostate cancer tissues also support the notionthat p53 mutation in SCNC may lead to a higher miR-25 expres-sion that contributes to the higher expression of Aurora kinase A.

DiscussionProstatic small cell carcinoma is an underdiagnosed entity

because patients with widely metastatic disease after hormonaltherapy usually do not undergo biopsy or resection for histologic

diagnosis. Its incidence is expected to rise after the recent approvalof super-blockers of AR signaling pathway such as abiraterone andenzalutamide. In addition to a lack of effective therapy, themolecularmechanisms driving the development of SCNC remainunclear.

We recently showed that p53 mutation may be a criticalmolecular responsible for the aggressive behavior of SCNC(23), and the study from Rubin's group reported gene amplifi-cation and overexpression of MYCN and Aurora kinase A in thesetumors (4). Aurora kinase A is an evolutionarily conserved serine/threonine kinase critical for mitotic regulation (23). It can phos-phorylate multiple mitosis-associated proteins (e.g., Tacc andNdel1), thus modulating their activities (24, 25) and orchestrat-ing centrosome maturation, spindle assembly, and mitotic entry.Aurora kinase A can also regulate protein translation throughCPEB phosphorylation (26, 27). Thus, Aurora kinase A playsimportant roles in cell proliferation and has been considered apotential therapeutic target for prostatic SCNC. In addition, it hasbeen also reported thatwild-type p53 suppresses the expression ofmiR-25 (12), and mutant p53 or loss of p53 function resulted inincreased miR-25 expression. The relationship between mutant

A B

C D

P53

FBXW7

AURKA

GAPDH

FBX

W7/

GA

PDH

AU

RK

A/G

APD

H

FBXW7

FBXW7 3′ UTR 1623bp

luc-FBXW7 3′UTR +miR Ctrl

luc-FBXW7 3′UTR +miR-25

luc-FBXW7 3′UTRm1,2+ miR Ctrl

luc-FBXW7 3′UTRm1,2+ miR-25

mutant FBXW7 3′ UTRCMV Luc.

Rel

ativ

e lu

cife

rase

act

ivity

CMV Luc.

AURKA

GAPDH

NE1.8

NE1.81.2

1

0.8

0.6

0.4

0.2

0

1 2 3LNCaP

4

1 2 3 4

1 2 3 4

MW(kD)100

70

40

55

35

MW(kD)100

70

40

35

siCtrl

sip53

sip53

+miR

inhib

itor N

eg C

trl

miR-C

trl

miR-25

siP53

+miR-25 i

nhibito

r

1.50

1.00

0.50

0.00

2.00

1.00

0.00

Figure 3.miR-25–regulated Aurora kinase A isdependent on loss of p53 function. A,Western blot analysis showing thatmiR-25 inhibitor eliminates the effectof p53 knockdown on regulation ofFBXW7 and AURKA. B, quantificationof the densitometry for Western blotbands using ImageJ 1.47t. C, theeffects of miR-25 on expressions ofAUKUA and FBXW7 proteins. Controlor miR-25 mimic at 100 nmol/L wastransfected into NE1.8 cells, a variantof LNCaP cells that resemble NE cells.Aurora kinase A and FBXW7 wereexamined via immunoblot analysis.D, a graphic representation forexpression plasmids containing fireflyluciferase with wild-type 30 UTR ofhuman FBXW7 and the mutant 30UTRwithout the target sequence formiR-25 (top). These plasmids weretransfected into NE1.8 cells togetherwith control or miR-25 mimic,luciferase activities weremeasured 48hours later and plotted (bottom).Three independent experiments wereperformed, and results are presentedas mean � SD.

Li et al.





p53 and overexpression of Aurora kinase A in SCNC is thus quiteinteresting. In present study, we show that expression of mutantp53 protein leads to enhanced expression of Aurora kinase Ain prostate cancer cells, and this was most likely mediatedby increased miR-25 expression and decreased expression ofFBXW7 subsequent to p53 mutation.

miR-25 is a well-studied oncogenic miRNA. It is 22 nucleotideslong, localized in the minichromosome maintenance protein-7(MCM7) gene, and transcribed as part of the miR-106b�25 poly-

cistron. It is overexpressed in several human cancers, includingpediatric brain tumors (28), gastric adenocarcinoma (29), EGFR-positive lung adenocarcinoma (30), and prostate carcinoma (31),and has been reported to target different regulators of the apo-ptotic pathway, such as BIM (32), PTEN (31), and TRAIL (33).miR-25 also affects Ca2þ homeostasis by regulating mitochondriacalcium efflux through targeting the mitochondria calcium uni-porter (34), causing a strong decrease in mitochondrial Ca2þ

uptake and, likely, conferring resistance to Ca2þ-dependent apo-ptotic stimuli. We found that p53 mutation–induced miR-25overexpression downregulates the expression of ubiquitin E3ligase FBXW7. FBXW7 is a potent ubiquitin E3 ligase that candegrade Aurora kinase A (35), lower level of FBXW7 thus leadsto increased protein level of Aurora kinase A. In PC-3 and NE1.8cells, we noticed that transfection with miR-25 reduced mRNAexpressions of Aurora kinase A, but significantly increased proteinlevels of Aurora kinase A which can be affected by exposure tocycloheximide, a protein synthesis inhibitor (Supplementary Fig.S4), suggesting that the elevatedprotein level of Aurora kinaseA inthese cells was caused by the FBXW7-induced blockage of Aurorakinase A degradation. Inactivation of FBXW7 has also beennoticed to be critical for the proliferation of leukemic stem cells,and contributes to the development of leukemia (36, 37). Thus,our results suggest a potential signaling pathway that howmutantp53 regulates level of Aurora kinase A in prostate cancer cells that

SCNC

mutationp53

miR-25 FBXW7

MYCN

Genomic amplification

AURKA

Figure 5.Adiagramdepictingmultiple pathways that lead to overexpression of Aurorakinase A in human prostatic small cell carcinoma.

A B

C

D

E

P53

Cas

e1C

ase2

AURKA AURKA

AU

RK

A

DAPI

P53/AURKACoexpression

P53 P53 MERGE

PCa SCNC

PCaQuick scoreFrequency

0–100100–200200–300Total

36201470

51.428.620.0100.0

11911

9.19.1

81.8100.0

Percent Frequency PercentSCNC

Quick score for PCa and SCNC

Adeno. PCaSCNC

AURKA

: 6/12

IHC+ : 8/12IHC+ : 9/12

20018016014012010080604020

0Rat

io o

f miR

-25/

miR

-191

Figure 4.Expressions of p53, Aurora kinaseA, and miR-25 in human prostaticsmall cell carcinoma. A,representative images show theIHC staining with anti-p53 and anti–Aurora kinase A antibodies in humanprostatic small cell carcinoma B,representative images showing thecoexpression of nuclear p53 andAurora kinase A in these humanprostatic small cell carcinoma usingimmunofluorescence doublestaining. C, miR-25 was quantitatedin small cell carcinoma oradenocarcinoma into box-and-whisker plots, the ratio of themiR-25over miR-191 as an internal controlwas plotted. D, differentialexpression of Aurora kinase A inhuman prostatic adenocarcinoma orsmall cell carcinoma. Representativeimage showing the results for IHCstaining with antibody for Aurorakinase A in these different types ofprostate cancers (PCa). E,semiquantitation of AURKA IHCstaining was performed using theQuick score method.






may be correlated to rapid proliferation and aggressive behaviorof prostatic SCNC.

Although our data suggest the presence of a linear pathway ofp53 ! miR-25 ! FBXW7 ! Aurora kinase A in SCNC, there arelikely other important players in the pathogenesis of prostaticSCNC. In a significant number of SCNC cases, Aurora kinase Aoverexpression is associated with gene amplification, whichinvolves different genetic events. Of note, Rubin's group hasshown thatMYCN is also amplified andoverexpressed inprostaticSCNC (4). Because expression of FBXW7 can also cause degra-dation of MYC, it would be interesting to study if p53 mutationalso leads to MYC overexpression. In addition, Collins' groupreported that downregulation of the REST transcription complexmay lead to the development of SCNC (38). These diverse find-ings suggest that the pathogenesis of SCNC is a complex processthat may involve different players and multiple signaling path-ways (Fig. 5).

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

Authors' ContributionsConception and design: Y. Sun, X. Chen, C. Liang, J. HuangDevelopment of methodology: Z. Li, Y. Sun, C. Liang, J. HuangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Z. Li, Y. Sun, X. Chen, J. SquiresAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Z. Li, Y. Sun, B. Nowroozizadeh, J. Huang

Writing, review, and/or revision of the manuscript: Z. Li, Y. Sun, X. Chen,J. Squires, J. HuangAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): Z. LiStudy supervision: J. Huang

AcknowledgmentsThe authors thank Dr. Ming-Fong Lin for sharing NE1.8 cells.

Grant SupportThis research is supported by a Stand Up to Cancer–Prostate Cancer

Foundation–Prostate Dream Team Translational Cancer Research Grant(SU2C-AACR-DT0812). This research grant is made possible by the generoussupport of the Movember Foundation. Stand Up To Cancer is a program ofthe Entertainment Industry Foundation administered by the AmericanAssociation for Cancer Research. J. Huang is supported by UCLA SPORE inProstate Cancer (PI: Reiter), Department of Defense Prostate CancerResearch Program W81XWH-11-1-0227 (PI: Huang), W81XWH-12-1-0206(PI: L. Wu), NIH 1R01CA158627-01 (PI: Marks), 1R01CA172603-01A1 (PI:Jiaoti Huang), Prostate Cancer Foundation Honorable A. David MazzoneSpecial Challenge Award (PI: Robert Reiter), and Stand-up-to-Cancer DreamTeam Award (PI: Small and Witte).

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

Received May 19, 2014; revised November 3, 2014; accepted December 7,2014; published OnlineFirst December 15, 2014.

References1. Abrahamsson PA, Falkmer S, Falt K, Grimelius L. The course of neuroen-

docrine differentiation in prostatic carcinomas. An immunohistochemicalstudy testing chromogranin A as an "endocrine marker". Pathol Res Pract1989;185:373–80.

2. Abrahamsson PA, Wadstrom LB, Alumets J, Falkmer S, Grimelius L.Peptide-hormone- and serotonin-immunoreactive tumour cells in carci-noma of the prostate. Pathol Res Pract 1987;182:298–307.

3. Chen H, Sun Y, Wu C, Magyar CE, Li X, Cheng L, et al. Pathogenesis ofprostatic small cell carcinoma involves the inactivationof the P53pathway.Endocr Relat Cancer 2012;19:321–31.

4. Beltran H, Rickman DS, Park K, Chae SS, Sboner A, MacDonald TY,et al. Molecular characterization of neuroendocrine prostatecancer and identification of new drug targets. Cancer Discov 2011;1:487–95.

5. Zhang XQ, Kondrikov D, Yuan TC, Lin FF, Hansen J, Lin MF.Receptor protein tyrosine phosphatase alpha signaling is involved inandrogen depletion-induced neuroendocrine differentiation of andro-gen-sensitive LNCaP human prostate cancer cells. Oncogene 2003;22:6704–16.

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

7. Xin L, Teitell MA, Lawson DA, Kwon A, Mellinghoff IK, Witte ON.Progression of prostate cancer by synergy of AKT with genotropic andnongenotropic actions of the androgen receptor. Proc Natl Acad Sci U S A2006;103:7789–94.

8. Tiscornia G, Singer O, Verma IM. Production and purification of lentiviralvectors. Nat Protoc 2006;1:241–5.

9. Livak KJ, Schmittgen TD. Analysis of relative gene expression data usingreal-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods2001;25:402–8.

10. Detre S, Saclani Jotti G, Dowsett M. A "quickscore" method for immuno-histochemical semiquantitation: validation for oestrogen receptor inbreast carcinomas. J Clin Pathol 1995;48:876–8.

11. Suzuki HI, Yamagata K, Sugimoto K, Iwamoto T, Kato S, Miyazono K.Modulation of microRNA processing by p53. Nature 2009;460:529–33.

12. Suh SS, Yoo JY, Nuovo GJ, Jeon YJ, Kim S, Lee TJ, et al. MicroRNAs/TP53feedback circuitry in glioblastoma multiforme. Proc Natl Acad Sci U S A2012;109:5316–21.

13. Tai S, Sun Y, Squires JM, Zhang H, OhWK, Liang CZ, et al. PC3 is a cell linecharacteristic of prostatic small cell carcinoma. Prostate 2011;71:1668–79.

14. LuD,DavisMP, Abreu-Goodger C,WangW,Campos LS, Siede J, et al.MiR-25 regulates Wwp2 and Fbxw7 and promotes reprogramming of mousefibroblast cells to iPSCs. PLoS One 2012;7:e40938.

15. Otto T, Horn S, Brockmann M, Eilers U, Schuttrumpf L, Popov N, et al.Stabilization of N-Myc is a critical function of Aurora A in humanneuroblastoma. Cancer Cell 2009;15:67–78.

16. Yada M, Hatakeyama S, Kamura T, Nishiyama M, Tsunematsu R, Imaki H,et al. Phosphorylation-dependent degradation of c-Myc is mediated by theF-box protein Fbw7. EMBO J 2004;23:2116–25.

17. Welcker M, Orian A, Jin J, Grim JE, Harper JW, Eisenman RN, et al. TheFbw7 tumor suppressor regulates glycogen synthase kinase 3 phosphor-ylation-dependent c-Myc protein degradation. Proc Natl Acad Sci U S A2004;101:9085–90.

18. Koepp DM, Schaefer LK, Ye X, Keyomarsi K, Chu C, Harper JW, et al.Phosphorylation-dependent ubiquitination of cyclin E by the SCFFbw7ubiquitin ligase. Science 2001;294:173–7.

19. Strohmaier H, Spruck CH, Kaiser P, Won KA, Sangfelt O, Reed SI. HumanF-box protein hCdc4 targets cyclin E for proteolysis and is mutated in abreast cancer cell line. Nature 2001;413:316–22.

20. Wang Z, InuzukaH, Zhong J,Wan L, Fukushima H, Sarkar FH, et al. Tumorsuppressor functions of FBW7 in cancer development and progression.FEBS Lett 2012;586:1409–18.

21. Mao JH, Perez-Losada J, WuD, Delrosario R, Tsunematsu R, Nakayama KI,et al. Fbxw7/Cdc4 is a p53-dependent, haploinsufficient tumour suppres-sor gene. Nature 2004;432:775–9.

22. Mosquera JM, Beltran H, Park K,MacDonald TY, Robinson BD, Tagawa ST,et al. Concurrent AURKA andMYCN gene amplifications are harbingers of


Li et al.




lethal treatment-related neuroendocrine prostate cancer. Neoplasia 2013;15:1–10.

23. Barr AR, Gergely F. Aurora-A: themaker and breaker of spindle poles. J CellSci 2007;120:2987–96.

24. Barros TP, Kinoshita K, Hyman AA, Raff JW. Aurora A activates D-TACC-Msps complexes exclusively at centrosomes to stabilize centrosomalmicro-tubules. J Cell Biol 2005;170:1039–46.

25. Mori D, Yano Y, Toyo-oka K, Yoshida N, Yamada M, Muramatsu M, et al.NDEL1 phosphorylation by Aurora-A kinase is essential for centrosomalmaturation, separation, and TACC3 recruitment. Mol Cell Biol 2007;27:352–67.

26. Mendez R, Hake LE, Andresson T, Littlepage LE, Ruderman JV, Richter JD.Phosphorylation of CPE binding factor by Eg2 regulates translation of c-mos mRNA. Nature 2000;404:302–7.

27. Mendez R, Murthy KG, Ryan K, Manley JL, Richter JD. Phosphorylation ofCPEB by Eg2 mediates the recruitment of CPSF into an active cytoplasmicpolyadenylation complex. Mol Cell 2000;6:1253–9.

28. Birks DK, Barton VN, Donson AM, Handler MH, Vibhakar R, Foreman NK.Survey of MicroRNA expression in pediatric brain tumors. Pediatr BloodCancer 2011;56:211–6.

29. Petrocca F, Visone R,OnelliMR, ShahMH,NicolosoMS, deMartino I, et al.E2F1-regulated microRNAs impair TGFbeta-dependent cell-cycle arrestand apoptosis in gastric cancer. Cancer Cell 2008;13:272–86.

30. Dacic S, Kelly L, Shuai Y, Nikiforova MN. miRNA expression profiling oflung adenocarcinomas: correlation with mutational status. Mod Pathol2010;23:1577–82.

31. Poliseno L, Salmena L, Riccardi L, Fornari A, Song MS, Hobbs RM, et al.Identification of themiR-106b�25microRNA cluster as a proto-oncogenicPTEN-targeting intron that cooperates with its host gene MCM7 in trans-formation. Sci Signal 2010;3:ra29.

32. ZhangH, Zuo Z, Lu X,Wang L,WangH, Zhu Z.MiR-25 regulates apoptosisby targeting Bim in human ovarian cancer. Oncol Rep 2012;27:594–8.

33. Razumilava N, Bronk SF, Smoot RL, Fingas CD, Werneburg NW, RobertsLR, et al. miR-25 targets TNF-related apoptosis inducing ligand (TRAIL)death receptor-4 and promotes apoptosis resistance in cholangiocarci-noma. Hepatology 2012;55:465–75.

34. Marchi S, Lupini L, Patergnani S, Rimessi A, Missiroli S, Bonora M, et al.Downregulation of the mitochondrial calcium uniporter by cancer-relatedmiR-25. Curr Biol 2013;23:58–63.

35. Fujii Y, YadaM,NishiyamaM, Kamura T, TakahashiH, Tsunematsu R, et al.Fbxw7 contributes to tumor suppression by targetingmultiple proteins forubiquitin-dependent degradation. Cancer Sci 2006;97:729–36.

36. King B, Trimarchi T, Reavie L, Xu L, Mullenders J, Ntziachristos P, et al. Theubiquitin ligase FBXW7 modulates leukemia-initiating cell activity byregulating MYC stability. Cell 2013;153:1552–66.

37. Takeishi S, Matsumoto A, Onoyama I, Naka K, Hirao A, Nakayama KI.Ablation of Fbxw7 eliminates leukemia-initiating cells by preventingquiescence. Cancer Cell 2013;23:347–61.

38. Lapuk AV, Wu C, Wyatt AW, McPherson A, McConeghy BJ, Brahmbhatt S,et al. From sequence tomolecular pathology, and amechanism driving theneuroendocrine phenotype in prostate cancer. The Journal of pathology2012;227:286–97.






2015;13:584-591. Published OnlineFirst December 15, 2014.Mol Cancer Res Zhen Li, Yin Sun, Xufeng Chen, et al. FBXW7 in Prostatic Small Cell Neuroendocrine Carcinoma

andmiR-25p53 Mutation Directs AURKA Overexpression via

Updated version

10.1158/1541-7786.MCR-14-0277-Tdoi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2014/12/16/1541-7786.MCR-14-0277-T.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/13/3/584.full#ref-list-1

This article cites 38 articles, 13 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/13/3/584.full#related-urls

This article has been cited by 4 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://mcr.aacrjournals.org/content/13/3/584To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-14-0277-T

http://mcr.aacrjournals.org/content/suppl/2014/12/16/1541-7786.MCR-14-0277-T.DC1

http://mcr.aacrjournals.org/content/13/3/584.full#ref-list-1

http://mcr.aacrjournals.org/content/13/3/584.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/13/3/584


p53 Mutation Directs AURKA Overexpression via miR-25 and ... · miR-25 and FBXW7 in Prostatic Small...

Documents

Transcript of p53 Mutation Directs AURKA Overexpression via miR-25 and ... · miR-25 and FBXW7 in Prostatic Small...