p53 gene mutational rate, Gleason score, and BK virus infection in prostate adenocarcinoma: Is there a correlation?

Journal of Medical Virology 80:2100–2107 (2008)

p53 Gene Mutational Rate, Gleason Score, andBK Virus Infection in Prostate Adenocarcinoma:Is There a Correlation?

Giuseppe Russo,1,2 Elena Anzivino,3 Daniela Fioriti,4 Monica Mischitelli,3 Anna Bellizzi,3

Antonio Giordano,1,5 Anamaria Autran-Gomez,4 Franco Di Monaco,4 Franco Di Silverio,4

Patrizio Sale,6,7 Laura Di Prospero,7 and Valeria Pietropaolo1,3*1Sbarro Institute for Cancer Research and Molecular Medicine, Center for Biotechnology,College of Science and Technology, Temple University, Philadelphia, Pennsylvania2DISI-Department of Computer and Information Sciences, University of Genoa, Italy3Department of Public Health Sciences, ‘‘La Sapienza’’ University, Rome, Italy4Department of Urology, ‘‘La Sapienza’’ University, Rome, Italy5Department of Human Pathology and Oncology, University of Siena, Siena, Italy6IRCCS San Raffaele ‘‘La Pisana’’, Rome, Italy7Department of Experimental Medicine and Pathology, ‘‘La Sapienza’’ University, Rome, Italy

Prostate cancer represents the second leadingcause of cancer deaths in Western countries. Viralinfections could play a role in prostate carcino-genesis. Human polyomavirus BK (BKV) is apossible candidate because of its transformingproperties. In this study, BKV sequences in urine,blood, fresh, and paraffin-embedded prostatecancer samples from 26 patients were searchedusing Q-PCR analysis. T antigen (TAg) and p53localization in neoplastic cells were evaluated byimmunohistochemical analysis. Also, the pres-ence of mutations in 5–9 exons of p53 gene wasanalyzed. Results showed that BKV-DNA wasfound in urine (54%), plasma (31%), and in freshprostate cancer specimens (85%). The analysis ofp53 gene evidenced several mutations in highGleason patients, according to tumor advancedstage. Immunohistochemical analysis resultsevidenced the localization of p53 and TAg intocytoplasm, whereas in TAg-negative tumors, p53was nuclear. This study suggests that BKV acts ascofactor in the pathogenesis of prostate cancer.These observations emphasize previous studiesregarding the cellular pathways that may bederegulated by BKV. J. Med. Virol. 80:2100–2107, 2008. � 2008 Wiley-Liss, Inc.

KEY WORDS: viral neoplastic transformation;BKV-TAg; quantitative PCR;prostate cancer; p53

INTRODUCTION

Prostate cancer is the most commonly diagnosedmalignancy and the second leading cause of cancer

death in American and Italian men [Jemal et al., 2005;Tumori in Italia – Rapporto, 2006]. The molecularpathology of prostate cancer is complex [Hughes et al.,2005]. Several factors are related to the developmentand progression of prostate cancer: Age is the mostsignificant risk factor, but several genes also play arelevant role in carcinogenesis of the prostate [Abate-Shen and Shen, 2000]. Sporadic prostate cancer isstrongly influenced by polymorphisms associated withincreased risk of prostate cancer (androgen receptor,growth factors, invasion and metastasis genes, and cellcycle regulators). Tumor suppressor genes are mostlikely involved in carcinogenesis of the prostate.

Known as the ‘‘guardian of the genome,’’ the tumorsuppressor p53 plays a critical role in controlling the cellcycle. As a transcription factor, it mediates the expres-sion of genes involved in growth, differentiation, andproliferation. Its role in prostate cancer is not very clear.Although the mutation rate of this gene in prostatecancer is low, p53 inactivation has been implicated intumor progression [Karayi and Markham, 2004; Hugheset al., 2005]. Single point mutations (missense muta-tions) abrogate the p53 suppressor function and con-tribute to the transformed phenotype [Dong, 2006]. Infact, mutated alleles of p53 encode a stable protein thataccumulates to high levels in tumor cell nuclei, whereas

Grant sponsor: Ministero dell’Universita e della Ricerca.

*Correspondence to: Valeria Pietropaolo, PhD, Faculty ofMedicine, Department of Public Health Sciences, ‘‘La Sapienza’’University, P.le Aldo Moro, 5, 00185 Rome, Italy.E-mail: [email protected]

Accepted 29 July 2008

DOI 10.1002/jmv.21312

Published online in Wiley InterScience(www.interscience.wiley.com)

� 2008 WILEY-LISS, INC.

wild-type p53 degrades quickly in normal cells. Usingimmunohistochemistry and mutation analysis, abnor-mal nuclear p53 accumulation and p53 mutations havebeen observed in prostate cancer [Dong, 2006]. It isknown that mutations of p53 are rare in primaryprostate cancer, but are more common in prostatecancer at a higher tumor stage, higher tumor grade,metastases, or androgen-independent tumors [Dong,2006]. It seems that exons 7 and 8 are more susceptibleto mutation in prostate cancer [Web-Site: IARC TP53Mutation Database, 2006].

The basic molecular mechanisms regulating thedevelopment and progression of prostate cancer arepoorly understood [Varambally et al., 2002; Rubin et al.,2004]. Researchers speculated that numerous factors(i.e., infectious agents) might play a role in the patho-genesis and/or progression of prostate cancer. Thehuman polyomavirus BK (BKV) is a good candidatebecause it naturally infects humans, it is usuallyacquired early in life, and almost 90% of adults haveantibodies to BKV [Hirsch and Steiger, 2003; Lundstigand Dillner, 2006]. However, its etiological role inhuman tumors is debated despite several reportsdemonstrating the presence and expression of viralsequences in neoplastic cells [Das et al., 2004]. BKV wasfirst isolated from the urine of a renal transplant patient[Gardner et al., 1971] and it infects persistently thehuman population in early childhood [Dorries et al.,1994]. The virus resides in the kidneys in a latent orpersistent state, but it can be reactivated by immuno-suppression of the host and it is associated withhemorrhagic cystitis and polyomavirus nephropathy inrenal transplants [Hirsch and Steiger, 2003]. Variousin vitro experiments demonstrated the oncogenic poten-tial of BKV [Imperiale, 2001; Tognon et al., 2003] andviral DNA was detected in vivo in several human tumorssuch as the urinary tract, brain, pancreatic islet, lung,the eye, carcinoma of the liver, and Kaposi’s sarcoma[Monini et al., 1995; Reploeg et al., 2001; Tognon et al.,2003].

The BKV genome is divided into regulatory, early, andlate regions, and encodes six major proteins [Fioritiet al., 2005]. The early proteins are the large tumorantigen (TAg), the main regulatory protein, and thesupporting small tumor antigen (tAg). TAg regulatesviral DNA replication and viral gene expression and caninfluence gene expression of activated and basal cellulartranscription factors. TAg promotes oncogenic trans-formation of cells inactivating tumor suppressors,including p53 and pRb proteins [Tognon et al., 2003;Das et al., 2004; Fioriti et al., 2005]. TAg binds p53 from92 to 292 amino acids (exons 4–8) [Lilyestrom et al.,2006].

The presence of BKV sequences was sought in urine,blood, and fresh prostate cancer specimens from radicalprostatectomies by a quantitative polymerase chainreaction (Q-PCR). As a control group, urine, blood, fresh,and paraffin-embedded tissue samples from 12 patientswith histological diagnosis of benign prostatic hyper-plasia were examined. In addition, the localization of

BKV-TAg and tumor suppressor p53 in neoplastic cellswas examined by immunohistochemistry with anti-bodies specific to TAg or p53. Finally, p53 sequencinganalysis of the region with more then 98% of mutationspresent (exon 5–9, specific DNA binding domains) wascarried out to understand if p53 mutation might becorrelated with viral infection and/or progression of thecancer.

MATERIALS AND METHODS

Patient Population

Twenty-six patients with prostate cancer wereenrolled in the present study. The samples examined(urine, blood, fresh, and paraffin-embedded tissues)were obtained from patients admitted to the ‘‘Umberto I’’Hospital of Rome Italy, with a diagnosis of clinically non-metastatic prostate adenocarcinoma and underwentradical prostatectomy. The median age of the patientswas 66 (range, 53–79). The patients were enrolled inthis study from January 2006 to January 2007 accordingto the following criteria:

(1) no previous hormonal or radiation therapy;(2) no previous surgery on the prostate gland;(3) histologically proven prostate cancer by biopsy and

confirmed at radical prostatectomy.

All 26 patients had a biopsy proven clinically T2-T3NoMo prostate adenocarcinoma, as determined bydigital rectal examination, transrectal ultrasonogra-phy, bone scan and computer tomography (CT), andserum levels of the prostate specific antigen (PSA). Thetumors chosen for this study were selected on the basis ofseveral criteria: histopathological diagnosis of thetumor, tumor size, grade, stage, androgen receptorstatus, and family history. Pathological tumor stage,after radical prostatectomy, was assigned according tothe 2002 TNM classification of the International UnionAgainst Cancer [UICC, 2002]. Tumor grade wasdescribed at radical prostatectomy according to theGleason grading system. For each patient, informedconsent, clinical-pathological and follow-up data areavailable in a computerized registry database.

Twelve patients with histological diagnosis of benignprostatic hyperplasia were used as control group. Thesamples examined (urine, blood, fresh, and paraffin-embedded tissue samples) were from patients admittedto the ‘‘Umberto I’’ Hospital of Rome, Italy with a medianage of 69 (range, 60–78) and without a history ofneoplastic diseases.

Clinical Specimens Processing

Urine. One milliliter of urine sample was incubatedin lysis buffer and proteinase K (200 mg/ml). DNAextraction was performed by the DNeasy1 Tissue Kit(QIAGEN, S.p.A, Italy) according to the manufacturer’sinstructions. One microgram of total purified DNA wasused for Q-PCR.

J. Med. Virol. DOI 10.1002/jmv

BKV Infection in Prostate Cancer Patients 2101

Blood. Blood samples in EDTA were centrifuged at1,376 g/sec for 10 min and 200 ml of plasma wereincubated in lysis buffer and proteinase K (200 mg/ml).DNA extraction was performed by the QIAmp1 DNABlood Kit (QIAGEN) according to the manufacturer’sinstructions. One microgram of purified DNA was usedfor Q-PCR.

Biopsy. Fresh and paraffin-embedded prostate can-cer resections were obtained from each patient. Inparallel, as controls, fresh, and paraffin-embeddedbenign prostatic hyperplasia resections were analyzed.About 25 mg of each fresh sample were incubated in lysisbuffer and proteinase K (200 mg/ml). DNA extractionwas performed by the DNeasy1 Tissue Kit (QIAGEN)according to the manufacturer’s instructions. Onemicrogram of total purified DNA was used for Q-PCR.

Immunohistochemistry was carried out using about25 mg of paraffin-embedded prostate cancer and benignprostatic hyperplasia resections. Samples were depar-affinized, rehydrated, and subjected to high temper-ature antigen retrieval in 10 mM sodium citrate buffer(pH 6.0). Endogenous peroxidase activity was blocked by3% H2O2.

Immunohistochemistry

Immunohistochemical testing was performed onparaffin embedded sections using specific antibodiesagainst BKV-TAg and p53 proteins. The primary anti-bodies used were the mouse monoclonal antibody forBKV Large TAg (PAb416, 1:200 dilution in blockingbuffer, Novocastra Laboratories Ltd., Newcastle uponTyne, UK) and the mouse monoclonal antibody for p53protein (p53Ab-6, Clone DO-1; 1:20 dilution in blockingbuffer, Lab Vision, Fremont, CA). The incubation wascarried out for 1 hr at room temperature; the secondaryantibody used for both TAg and p53 proteins wasbiotinylated goat anti-mouse IgG. All slides wereprocessed using preformed horseradish peroxidase-conjugated streptavidin (Biogenex, San Ramon, CA).The immunoreaction product was revealed using amino-ethylcarbazole (AEC) and 0.01% H2O2 (Biogenex,San Ramon, CA). Negative control staining was carriedout by omitting the primary antibody. Sections werecounterstained in Mayer’s acid hemalum and analyzed.

Quantitative PCR for BKV. Urine, blood, freshprostate cancer, and benign prostatic hyperplasia speci-mens were tested using Q-PCR for detection andquantitation of BKV-DNA. Q-PCR assay was performedusing 7300 Real Time PCR System (AB AppliedBiosystems, Foster City, CA). PCR amplifications wererun in a reaction volume of 20 ml containing 5 ml of theDNA sample, Q-Amplimaster (reaction mix for quanti-tative amplification), Q-Amplimix (forward and reverseprimers), and finally Q-Ampliprobe (hydrolysis probes)(Nanogen Advanced Diagnostics, S.r.l., San Diego, CA).Thermal cycling was initiated with a first denaturationstep of 10 min at 958C, followed by 45 cycles of 958C for15 sec, 608C for 1 min, and 728C for 1 min, at the end ofwhich fluorescence was read. The amplification data

were analyzed with software provided by the manufac-turer. Standard curves for the quantitation of BKV wereconstructed using serial dilutions of a plasmid contain-ing target sequences for BKV (Large TAg). The plasmidconcentrations ranged from 102 to 105 plasmid copies ofBKV-DNA target. All patient samples were tested intriplicate and the number of BKV copies in each samplewas calculated from the standard curve. Data wereexpressed as copies of viral DNA per milliliter of sample.Standard precautions designed to prevent contamina-tion during Q-PCR were followed. A not template control(NTC) lane was included in each run. The b-globin genewas used as internal standard (Nanogen AdvancedDiagnostics, S.r.l., San Diego, CA). Q-PCR allows thedetection of 10 molecules of target sequences in 5 ml ofextracted DNA used in the reaction.

Mutational Analysis of p53

The mutational analysis of p53 was performed bySNPCaptureTMp53 Mutation Screening Kit (Panomics,Fremont, CA) and sequence analysis [Das et al., 2008].SNPCaptureTMp53 Mutation Screening Kit (Panomics)is a gel-based method designed to detect geneticmutations in coding regions between exons 5 and 9 ofthe human gene p53 through Holliday junction for-mation. The reason for this restriction is that morethan 98% of p53 mutations in human neoplasias arelocated in these exons. In addition, exons 5–9 (codons126 and 331 with 540 base pairs) contain DNAsequences that code for domains necessary forsequence-specific DNA binding. This assay includesprimers for sequence-specific detection, convenientpremixed reagents, control wild-type, and mutant p53templates. Control wild-type templates were designedon the basis of GenBank data. Mutant templates containmissense mutations localized in the following codons:175 (exon 5), 213 (exon 6), 249 (exon 7), and 273 (exons 8and 9).

Extracted DNA from biopsies (target DNA) and thep53 wild-type DNA, which functions as a reference, wereamplified in a thermal cycler. Amplification was per-formed by a preliminary Taq polymerase activation stepat 958C for 10 min, followed by 45 cycles of 15 sec at 948Cfor DNA denaturation, 23 sec at 588C for primersannealing, and 1 min at 728C for DNA extension. Fivemicroliters of target DNA (approximately 5 ng) was usedin 50 ml of PCR reaction. Amplified target DNA wasmixed with amplified reference DNA (p53 wild-typeDNA) and subjected to a 658C temperature for 30 min toform stable Holliday junction between mutated targetsand reference DNA. Holliday junctions were detected byelectrophoresis on 1% agarose gel stained with ethidiumbromide. All assays were carried out in a GeneAmpPCR System 9700 (AB Applied Biosystems). Althoughthe use of SNPCaptureTMp53 Mutation Screening Kitnot required a confirming sequencing analysis, a p53sequencing analysis was carried out according toDas et al. [2008] in order to detect other pointmutations not revealed by the commercial kit. After


2102 Russo et al.

gene amplification, PCR products were separated byagarose gel electrophoresis, extracted (Qiaquick gel ex-traction kit; QIAGEN S.p.A) and sequenced by auto-matic DNA sequencer (Applied Biosystem, mod. 370A),according to manufacturer’s specifications (AmplicycleKit, Applied Biosystem). Sequences were organized andanalyzed using the Genetic Computer Group sequenceanalysis software package.

Statistical Analysis

Descriptive statistics were used to characterize thepopulation. For the statistical analysis, patientswere classified on the basis of the pathological T stage inpT2 and pT3 patients (no pT4 cases were found), onthe basis of the radical prostatectomy Gleason score: highrisk 8(4þ 4), 9(5þ 4); medium risk 6(3þ 3) 7(3þ4), onthe basis of the serum PSA levels in �10.0 ng/ml and>10 ng/ml.

Spearman correlation coefficients were calculated tomeasure the association among PSA, Gleason score,clinical stage and presence of BKV-DNA with otherparameters of mutation differences in the parametersbetween groups were tested using ANOVA non-para-metric test and chi-square test. Gleason score, PSA andpresence of BKV-DNA, and mutations classificationswere analyzed. Univariate and multivariate (Coxproportional hazard method) analysis were also under-

taken. A 5% level of significance was used for allstatistical testing. A Sigma-Stat and Sigma-Plot 2-2(Jandel Scientific software, San Rafael, CA) programhave been used for all statistical analysis.

RESULTS

In this study, urine, blood, fresh, and paraffin-embedded tissue samples, taken from 26 patients withprostate cancer, were tested by Q-PCR. Adenocarci-noma specimens were graded by cellular content andtissue architecture of the biopsies according to theGleason score. The final Gleason score of the samplesexamined is shown in Table I and in Table II. As controlgroup, urine, blood, fresh, and paraffin-embedded tissuesamples, taken from 12 patients with benign prostatichyperplasia, were analyzed.

Urine samples showed that 14/26 (54%) patientswere positive for BKV-DNA, whereas 8/26 (31%)blood samples were positive for BKV-DNA (Table I).No control subject was found positive (data not shown).

Fresh adenocarcinoma prostate resection specimenswere examined for BKV-DNA and a p53 gene muta-tional analysis was performed on the same samples.Using Q-PCR, 22/26 (85%) patients were positive forBKV-DNA. The average of viral copies number wascalculated on positive patients belonging to the sameGleason score class. Seven patients with Gleason 9showed an average of 16,914 c/ml, whereas in six


TABLE I. Immunohistochemistry and Q-PCR

SampleGleason

score

ImmunohistochemistryQ-PCR biopsies

(c/ml)Q-PCR blood

(c/ml)Q-PCR urine

(c/ml)TAga p53a

1 9 (5þ 4) þ þ 19,600 � 5202 9 (5þ 4) þ þ 17,000 � �3 9 (5þ 4) þ þ 20,400 ND �4 9 (4þ 5) þ þ 15,400 � �5 9 (4þ 5) þ þ 15,000 � 101,5346 9 (4þ 5) þ þ 16,400 1,527 1387 9 (4þ 5) þ þ 14,600 � 122,6888 8 (5þ 3) þ þ 14,000 � �9 8 (5þ 3) þ þ 13,600 � �

10 8 (5þ 3) þ þ 14,000 � �11 8 (4þ 4) þ þ 12,400 227 136,98612 8 (4þ 4) þ þ 13,000 � 107,32013 8 (4þ 4) þ þ 12,800 � �14 7 (4þ 3) þ þ 11,000 ND 31415 7 (4þ 3) þ þ 10,000 � �16 7 (4þ 3) þ þ 10,400 1,679 72317 7 (4þ 3) þ þ 8,200 � �18 7 (4þ 3) þ þ 10,200 1,476 128,65719 7 (3þ 4) þ þ 7,400 1,623 114,25620 7 (3þ 4) þ þ 9,000 800 154,95421 7 (3þ 4) � þ (N) � � �22 7 (3þ 4) � þ (N) 600 1,542 150,00023 6 (3þ 3) � þ (N) � � �24 6 (3þ 3) � þ (N) 1,000 1,438 148,65225 6 (3þ 3) � þ (N) � � 5026 6 (3þ 3) � þ (N) � � �

Data obtained by means of Immunohistochemistry and Q-PCR performed on biological samples of patients affected by prostate cancer. Thesamples have been listed in groups based on Gleason score. ND: Not determined.a�: negative; þ: positive and cytoplasmic; þ (N): positive and nuclear.


patients with Gleason 8 an average of 13,300 c/ml wasfound and finally seven patients with Gleason 7 showedan average of 9,457 c/ml. Patient no. 22 was excludedsince its low copy number might introduce an extremevariability in the same group. Regarding patients withGleason 6 score, one patient had a low copy number ofBKV-DNA (1,000 c/ml). Negative patients had Gleason 7score (one patient) and Gleason 6 score (three patients)(Table I).

Regarding p53 mutational analysis, all specimens hadat least one mutated exon of p53. It was found thatcodons 249 (exon 7) and 273 (exon 8) were moresusceptible to mutation for all examined patients [Dong,2006; Web-Site: IARC TP53 Mutation Database, 2006].Exon 7 was always mutated in patients with Gleasongrade 9. Mutations in exons 5, 6, 8, and 9 were observedin 1/7 patients; 1/7 patients had mutation in exons 6and 8, 1/7 patients in exons 5, 8, and 9, 2/7 patientsshowed mutation in exons 5 and 6, and 2/7 patients hadmutation in exon 8 (Table II). In patients with Gleasongrade 8, exons 5 and 8 were mutated in 2/6 patients;mutations in 7 and 8 exons were found in 3/6 patients,and mutations in exons 6 and 7 were observed in 1/6

patients (Table II). Finally, Gleason grade 7 BKVþ

patients showed a different pattern of mutations(Table II).

Within four BKV-DNA negative patients, 1/4 patientswith Gleason grade 7 showed mutations in exons 5 and6, whereas Gleason 6 patients showed mutations in exon7 (one patient), 5 and 7 (one patient), and 8 (one patient)(Table II bold marked and Fig. 1). The results obtainedby p53 sequencing analysis [Das et al., 2008] not onlyconfirmed these data but also demonstrated newmutations, which are showed in italics in Table II.

Immunohistochemistry results showed the presenceof BKV-TAg protein on scattered cells of the prostatecancer glandular epithelium (Fig. 2A), whereas inbenign prostatic hyperplasia, BKV-TAg was notdetected (Fig. 2B). p53 and TAg localization wasobserved in the cytoplasm of the neoplastic glandularepithelial cells (Table I and Fig. 3A), whereas in the TAg-negative tumors, p53 was localized in the nucleus(Table I and Fig. 3B). Immunohistochemistry carriedout on control specimens showed low levels of p53 intothe glandular epithelial cells’ cytoplasm (data notshown).


TABLE II. p53 Mutational Analysis Performed on Adenocarcinoma Samples of Patients Affected by Prostate Cancer

No. Gleason score Exon 5 Exon 6 Exon 7 Exon 8 Exon 9

1 9 (5þ 4) Arg 175–HisHis 179–His

Arg 213–Arg Arg 249–Ser Arg 273–His Ala 307–Val

2 9 (5þ 4) wt Arg 213–Arg Arg 249–SerArg 248–Pro

Arg 273–His wt

3 9 (5þ 4) Arg 175–His wt Arg 249–Ser Arg 273–His Gln 331–Leu4 9 (4þ 5) Arg 175–His Arg 213–Arg Arg 249–Ser wt wt5 9 (4þ 5) Arg 175–His Arg 213–Arg Arg 249–Ser wt wt6 9 (4þ 5) wt wt Arg 249–Ser Arg 273–His

Arg 282–Leuwt

7 9 (4þ 5) wt wt Arg 249–SerArg 248–Leu

Arg 273–His wt

8 8 (5þ 3) Arg 175–His wt wt Arg 273–His wt9 8 (5þ 3) Arg 175–His wt wt Arg 273–His wt

10 8 (5þ 3) wt wt Arg 249–SerArg 248–Arg

Arg 273–His wt

11 8 (4þ 4) wt wt Arg 249–Ser Arg 273–His wt12 8 (4þ 4) wt wt Arg 249–Ser Arg 273–His wt13 8 (4þ 4) wt Arg 213–Arg

Tyr 220–CysArg 249–Ser wt wt

14 7 (4þ 3) Arg 175–His wt Arg 249–Ser wt wt15 7 (4þ 3) wt Arg 213–Arg wt Arg 273–His wt16 7 (4þ 3) wt Arg 213–Arg wt Arg 273–His wt17 7 (4þ 3) Arg 175–His Arg 213–Arg Arg 249–Ser

Arg 248–Prowt wt

18 7 (4þ 3) wt wt wt Arg 273–His wt19 7 (3þ 4) Arg 175–His Arg 213–Arg wt wt wt20 7 (3þ 4) wt Arg 213–Arg

Arg 196–LeuArg 249–Ser wt wt

21 7 (3þ4) Arg 175–His Arg 213–Arg wt wt wt22 7 (3þ 4) wt Arg 213–Arg wt Arg 273–His wt23 6 (3þ3) wt wt Arg 249–Ser wt wt24 6 (3þ 3) wt wt Arg 249–Ser wt wt25 6 (3þ3) Arg 175–His wt Arg 249–Ser

Gly 245–Alawt wt

26 6 (3þ3) wt wt wt Arg 273–His wt

p53 mutational analysis performed using SNPCaptureTMp53 Mutation Screening Kit (Panomics) and the Das et al., protocol (mutationsexclusively found using the Das et al., protocol are marked in italics). BKV negative patients are marked in bold. The samples (N8) have been listedin groups based on the Gleason score.

2104 Russo et al.

Statistical analysis showed that there was no corre-lation between viral infection and p53 mutational rate inprostate cancer, whereas a statistically significantcorrelation between the Gleason score and the numberof p53 mutated exons was found, according to publisheddata [Dong, 2006].

Finally, descriptive statistics showed that there wasa significantly association between serum PSA levels,pT stage (r¼ 0.1983; P¼ 0.0012), and Gleason score(r¼ 0.1990; P¼ 0.001) (r¼Spearman coefficient).

DISCUSSION

Prostate cancer is the second leading cause of deathfrom cancer in western men [Jemal et al., 2005; Tumoriin Italia – Rapporto, 2006]. The basic molecularmechanisms regulating the development and progres-

sion of cancer of the prostate are poorly understood[Varambally et al., 2002; Rubin et al., 2004]. Never-theless, research demonstrated that a complex inter-action of multiple genes and environmental factorsoccurs [Hughes et al., 2005]. Recent insights gained fromdifferent studies suggested that infectious agents, suchas viruses, might play a key role in the pathogenesis and/or progression of cancer of the prostate [Das et al., 2004].

BKV is a possible candidate because it naturallyinfects humans and it is usually acquired early in lifewith antibody seroprevalence rate of almost 90% inadults. Since BKV is ubiquitous in the human popula-tion, there is speculation concerning the oncogenicpotential of the virus [Imperiale, 2000, 2001; Tognonet al., 2003]. Tumors of the urinary tract are the mostlogical target for an etiological association with BKV. A


Fig. 1. SNPCaptureTM p53 Mutation Screening Assay (Panomics). Holliday junctions are visualized byelectrophoresis on ethidium bromide-stained 1% agarose gel in 1� TAE buffer. M: low molecular weightmarker DNA (100 bp MBI Fermentas); (�): negative control; lanes 1–4: prostate adenocarcinoma samples;(þ): positive control; HJ: Holliday junction; DS: double stranded DNA amplicon. The sample 1 showsmutation in exons 8 and 9; the sample 2 shows mutation in exons 5, 7, 8, and 9; the sample 3 shows mutationin exons 6, 7, 8, and 9; the sample 4 shows mutation in exons 5 and 7.

Fig. 2. Expression of TAg in prostate sample immunostained withanti-TAg Ab. Tumoral prostate section derived from a BKV-infectedpatient corresponding to sample 5 (see Table I). A: The section showscuboidal epithelium cells labeled with anti-polyomavirus (BKV) (redarrow). B: The section shows a BHP section derived from a negativesubject for BKV infection. Original magnification 20�.

Fig. 3. Immunohistochemical analysis of p53 expression on prostatecancer sections from BKV-infected and uninfected patients. A:Prostate section from BKV-infected patient (sample 5, Table I) showingp53 expression. Positivity is present in the cytoplasm of cuboidalepithelium cells (black arrow). B: Prostate section from uninfectedpatient (sample 25, Table I) showing p53 expression. Strong positivityis present in the nucleus of cuboidal epithelium cells (black arrow).Original magnification 20�.


large number of studies demonstrated the presence ofBKV-DNA and BKV-TAg in prostate and bladdercarcinoma [Geetha et al., 2002; Weinreb et al., 2006;Das et al., 2008].

This study investigated the presence of BKV sequencesin urine, blood, and fresh prostate cancer specimens fromradical prostatectomies. Immunohistochemistry wascarried out to evaluate and to correlate the localizationof BKV-TAgand tumorsuppressorp53 in neoplastic cells.Finally, p53-sequencing analysis was carried out todetermine if gene mutation could be correlated with viralinfection and/or cancer progression. The results lead tothe hypothesis that: (1) BKV could induce cell trans-formation, suggesting a role of BKV for the pathogenesisof cancer; (2) BKV could promote development of tumors,suggesting that BKV must cooperate with cellularproteins for a complete ‘‘neoplastic phenotype’’ andprogression of metastatic disease.

In fresh prostate cancer specimens taken from 13patients with Gleason grade 8 and 9, Q-PCR detected ahigher number of BKV-DNA copies than other Gleasonclasses (16,914 and 13,300 vs. 9,457 and 1,000). In these13 patients, Q-PCR did not reveal the presence of BKV-DNA in seven urine and 10 blood samples. This resultsuggests the hypothesis that in these BKV negativepatients, viral infection could be elderly, whereasimmunosuppression could promote viral reactivationwithin six patients showing viruria. In two patientsviruria was associated to viremia.

In all Gleason grade 8 and 9 patients, immunohis-tochemistry showed cytoplasmic localization of bothBKV-TAg and p53, and different p53 mutated exonswere found according to tumor advanced stage.

In Gleason grade 7 patients, Q-PCR assay detected alower number of BKV-DNA copies than 8 and 9 Gleasongrades (9,457 c/ml) and the presence of BKV-DNA inurine (6/9 patients) and blood (5/9 patients). Five of ninepatients were positive to viral infection in both urine andblood samples. These data supported the hypothesisthat a primary BKV infection or reactivation took place.Regarding primary infection, the presence of viremiawas evidence of a recent BKV localization in tumor cells,whereas a BKV reactivation could be explained byimmunosuppression of the host.

In seven of nine positive patients with Gleason grade7, immunohistochemistry confirmed that both BKV-TAg and p53 were localized into the cytoplasm of theglandular epithelial cells. Regarding the two remainingpatients, BKV-TAg was not found and p53 was nuclear.Finally, in these patients, p53 mutational analysisevidenced a lower number of p53-mutated exons thanGleason grades 8 and 9, as expected in a medium tumorstage. Hence, it is suggested that BKV could play a keyrole in prostate cancer progression rather than inneoplasia onset in this Gleason class.

Regarding Gleason 6 class, only patient no. 24 hadBKV-DNA but the low copy number of BKV-DNA infresh prostate cancer specimens, the viruria, theviremia, and the nuclear p53 localization, suggest thatBKV infection was in act.

Statistical analysis confirmed a statistically signifi-cant correlation between Gleason score and the numberof p53 mutated exons [Dong, 2006].

Serum PSA level, clinical staging, and pathologicgrading and staging emerged as important factors in theevaluation of subjects with newly diagnosed prostatecancer. The descriptive statistic showed that there was asignificant association between serum PSA levels, pTstage, and Gleason score. Therefore, these parameterscan be used as markers, specific for the diagnosis ofcancer of the prostate.

In conclusion, the results are in agreement withseveral reports that confirm the presence of BKV-DNAand of its early protein TAg in numerous humanneoplasms [Sanchez-Chapado et al., 2003; Tognonet al., 2003; White and Khalili, 2004, 2006; Eash et al.,2006; Lee and Langhoff, 2006; Das et al., 2008]. Thesedata extend the findings that demonstrated the expres-sion of BKV-TAg in prostate tumors [Zambrano et al.,2002; Das et al., 2004]. It is confirmed that TAg waslocalized along with the tumor suppressor p53 in theglandular epithelial cells’ cytoplasm of prostatecancer sections, whereas in the absence of TAg, p53was nuclear. It is speculated that TAg drives prolifer-ation of these cells by sequestering and inactivating p53into the cytoplasm. This sequester blocks p53 interac-tion with other cellular proteins involved in cell cyclecontrol inhibiting cell cycle arrest or apoptosis. There-fore, it is possible to suggest that TAg represents themajor viral oncogenic protein [Khalili and Stoner, 2001].On the basis of results obtained, and since viral loads ofthe four Gleason classes have the same order ofmagnitude, it is concluded that the human polyomavi-rus BK acts as a cofactor in the pathogenesis of prostatecancer [Das et al., 2008]. These observations emphasizepublished studies on the possible cellular pathways thatmay be deregulated by BKV in the development ofprostate cancer. However, additional work will berequired to determine whether BKV acts only as acofactor or it plays a crucial role in the development ofprostate cancer.

REFERENCES

Abate-Shen C, Shen MM. 2000. Molecular genetics of prostate cancer.Genes Dev 14:2410–2434.

Das D, Shah RB, Imperiale MJ. 2004. Detection and expression ofhuman BK virus sequences in neoplastic prostate tissues. Oncogene23:7031–7046.

Das D, Wojno K, Imperiale MJ. 2008. BKV as a cofactor in the etiology ofprostate cancer in its early stages. J Virol 82:2705–2714.

Dong JT. 2006. Prevalent mutations in prostate cancer. J Cell Biochem97:443–447.

Dorries K, Vogel E, Gunther S, Czub S. 1994. Infection of humanpolyomaviruses JC and BK in peripheral blood leukocytes fromimmunocompetent individuals. Virology 198:59–70.

Eash S, Manley K, Gasparovic M, Querbes W, Atwood WJ. 2006. Thehuman polyomaviruses. Cell Mol Life Sci 63:865–876.

Fioriti D, Videtta M, Mischitelli M, Degener AM, Russo G, Giordano A,Pietropaolo V. 2005. The human polyomavirus BK: Potential role incancer. J Cell Physiol 204:402–406.

Gardner SD, Field AM, Coleman DV, Hulme B. 1971. New humanpapovavirus (BK) isolated from urine after renal transplantation.Lancet 1:1253–1257.


2106 Russo et al.

Geetha D, Tong BC, Racusen L, Markowitz JS, Westra WH. 2002.Bladder carcinoma in a transplant recipient: evidence to implicatethe BK human polyomavirus as a causal transforming agent.Transplantation 73:1933–1936.

Hirsch HH, Steiger J. 2003. Polyomavirus BK. Lancet Infect Dis 3:611–623.

Hughes C, Murphy A, Martin C, Sheils O, O’Leary J. 2005. Molecularpathology of prostate cancer. J Clin Pathol 58:673–684.

Imperiale MJ. 2000. The human polyomaviruses, BKV and JCV:molecular pathogenesis of acute disease and potential role incancer. Virology 267:1–7.

Imperiale MJ. 2001. Oncogenic transformation by the human poly-omaviruses. Oncogene 20:7917–7923.

Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A,Feuer EJ, Thun MJ. 2005. Cancer statistics. CA Cancer J Clin 55:10–30.

Karayi MK, Markham AF. 2004. Molecular biology of prostate cancer.Prostate Cancer Prostatic Dis 7:6–20.

Khalili K, Stoner GL. 2001. Human polyomaviruses: Molecular andclinical perspectives. Vol. 6. New York: Wiley-Liss, pp 439–442.

Lee W, Langhoff E. 2006. Polyomavirus in human cancer development.Adv Exp Med Biol 577:310–318.

Lilyestrom W, Klein MG, Zhang R, Joachimiak A, Chen XS. 2006.Crystal structure of SV40 large T-antigen bound to p53: interplaybetween a viral oncoprotein and a cellular tumor suppressor. GenesDev 20:2373–2382.

Lundstig A, Dillner J. 2006. Serological diagnosis of human poly-omavirus infection. Adv Exp Med Biol 577:96–101.

Monini P, Rotola A, Di Luca D, De Lellis L, Chiari E, Corallini A, CassaiE. 1995. DNA rearrangements impairing BK virus productiveinfection in urinary tract tumors. Virology 214:273–279.

Reploeg MD, Storch GA, Clifford DB. 2001. BK virus: A clinical review.Clin Infect Dis 33:191–202.

Rubin MA, Zerkowski MP, Camp RL, Kuefer R, Hofer MD, ChinnaiyanAM, Rimm DL. 2004. Quantitative determination of expression ofthe prostate cancer protein alpha-methylacyl-CoA racemase usingautomated quantitative analysis (AQUA): A novel paradigm forautomated and continuous biomarker measurements. Am J Pathol164:831–840.

Sanchez-Chapado M, Olmedilla G, Cabeza M, Donat E, Ruiz A. 2003.Prevalence of prostate cancer and prostatic intraepithelial neopla-sia in Caucasian Mediterranean males: An autopsy study. Prostate54:238–247.

Tognon M, Corallini A, Martini F, Negrini M, Barbanti-Brodano G.2003. Oncogenic transformation by BK virus and association withhuman tumors. Oncogene 22:5192–5200.

Tumori in Italia – Rapporto. 2006. Epidemiologia & Prevenzione 2: 70-71.

Varambally S, Dhanasekaran SM, Zhou M, Barrette TR, Kumar-SinhaC, Sanda MG, Ghosh D, Pienta KJ, Sewalt RG, Otte AP, Rubin MA,Chinnaiyan AM. 2002. The polycomb group protein EZH2 isinvolved in progression of prostate cancer. Nature 419:624–629.

Web-Site: IARC TP53 Mutation Database. 2006.

Weinreb DB, Desman GT, Amolat-Apiado MJ, Burstein DE, GodboldJH, Jr., Johnson EM. 2006. Polyoma virus infection is a prominentrisk factor for bladder carcinoma in immunocompetent individuals.Diagn cytopathol 34:201–203.

White MK, Khalili K. 2004. Polyomaviruses and human cancer:Molecular mechanisms underlying patterns of tumorigenesis.Virology 324:1–16.

White MK, Khalili K. 2006. Interaction of retinoblastoma proteinfamily members with large T-antigen of primate polyomaviruses.Oncogene 25:5286–5293.

Zambrano A, Kalantari M, Simoneau A, Jensen JL, Villarreal LP. 2002.Detection of human polyomaviruses and papillomaviruses inprostatic tissue reveals the prostate as a habitat for multiple viralinfections. Prostate 53:263–276.



Documents

p53 gene mutational rate, Gleason score, and BK virus infection in prostate adenocarcinoma: Is there a correlation?