IHaCF

O

M

R

C

Picwbthacnrdcw

FRp

H4

©A

Basic and Translational Science

mmunohistochemical Expression ofsp60 Correlates With Tumor Progressionnd Hormone Resistance in Prostate Cancer

arolina Castilla, Belén Congregado, José M. Conde, Rafael Medina,rancisco J. Torrubia, Miguel A. Japón, and Carmen Sáez

BJECTIVES To investigate the expression of Hsp60 protein in prostate cancer biopsy samples, and itsassociation with prognostic clinical parameters and hormone resistance and survival. Molecularchaperones are involved in protein folding, protein degradation, and protein trafficking amongsubcellular compartments.

ETHODS We selected 107 patients with localized and locally advanced prostate cancer at our hospital from1999 through 2004. We performed an analysis by western blot and immunohistochemistry onparaffin-embedded tissue sections. Clinical data were used to determine associations betweenimmunohistochemical expression of Hsp60 and tumor behavior.

ESULTS The expression level of Hsp60 was significantly increased in tumors with high Gleason score (P �.001). Hsp60 expression was also significantly associated with initial serum PSA levels (P � .01)and with the presence of lymph node metastasis (P � .003). In 50 locally advanced cancerstreated by androgen ablation we found an association between high Hsp60-expressing tumorsand an early onset of hormone refractory disease (P � .02) and reduced cancer-specific survival(P � .05).

ONCLUSIONS Hsp60 protein is overexpressed in poorly differentiated prostate cancers. Hsp60 expression isstrongly associated with prognostic clinical parameters, such as Gleason score, initial serum PSAlevels, and lymph node metastasis and with the onset of hormone-refractory disease and reducedcancer-specific survival. Identification of such markers could be of relevance in the clinical

management of prostate cancer. UROLOGY 76: 1017.e1–1017.e6, 2010. © 2010 Elsevier Inc.

aslwmpalchf

piltSbv

nm

rostate cancer is the leading cancer among men inindustrialized Western countries.1 Several mecha-nisms of prostate cancer progression have been

nvestigated, such as loss of chromosome 10q2 or alteredell-cycle regulatory genes, such as loss of p273 and p16,hich is rarely mutated in primary prostate carcinomas,ut it is in advanced metastatic disease.4 Mechanisms forhe development of androgen-independent phenotypeave not been fully unveiled. Although mutations in thendrogen receptor are not uncommon in prostate can-ers, they do not explain the androgen-independent phe-otype of many tumors that exhibit normal androgeneceptor expression.5 Prostate cancer is a heterogeneousisease with varying outcomes. For men diagnosed withlinically localized prostate cancer, definitive therapyith radical prostatectomy, external beam radiation ther-

rom the Instituto de Biomedicina de Sevilla (IBiS), Hospital Universitario Virgen delocío/CSIC/Universidad de Sevilla, Seville, Spain; and Department of Urology, De-artment of Pathology, Hospital Universitario Virgen del Rocío, Seville, Spain

Reprint requests: Carmen Sáez, M.D., Instituto de Biomedicina de Sevilla (IBiS),ospital Universitario Virgen del Rocío, Avenida Manuel Siurot S/n, Seville, Spain

a1013. E-mail: csaez@cica.es

Submitted: January 13, 2010, accepted (with revisions): May 23, 2010

2010 Elsevier Inc.ll Rights Reserved

py, or brachytherapy offers a high chance of cure. De-pite intensive efforts on early detection procedures, aarge proportion of prostate cancer patients still presentith advanced disease and there is no effective cure foren with advanced disease. Therapy against advanced

rostate cancer normally involves androgen ablation and,lthough initially effective, this therapy eventually fails,eading to a lethal hormone refractory disease.6 In thisontext, it is of interest the identification of markers thatelp to recognize patients prone to undergo therapeutic

ailure.Hsp60 belongs to a group of proteins involved in

rotein folding, protein degradation and protein traffick-ng among subcellular compartments, known as molecu-ar chaperones.7 These were initially discovered as pro-eins whose expression is induced upon heat treatment.ubsequently, they have been shown to be also inducedy a wide variety of stresses and to play critical prosur-ival functions.8

Hsp60 is abundant in most mammalian cells underormal conditions9 largely but not exclusively compart-entalized in mitochondria.10 Together with its associ-

ted chaperonin Hsp10, has a critical role in protein

0090-4295/10/$34.00 1017.e1doi:10.1016/j.urology.2010.05.045

fsspttsfacuecmcecwpibhwsrufad

ncnettmmb

M

PWtGlsgfsNictccHr

3AsPdtcbpmca

ITrltrcaiHawpGs3Taoescsc

WWs(ncsPjpw�baSwo

SAld

1

olding.11 Unlike Hsp27, -70, and -90 with obvious pro-urvival functions,12-14 Hsp60 appears to have both pro-urvival and prodeath functions. Existing data support aroapoptotic role via enhanced procaspase 3 matura-ion.15,16 By contrast, other studies suggest an antiapop-otic function involving sequestration of Bax17,18 ortabilization of survivin levels and restraint of p53unction.19 Recently, it has been suggested a dual pro-nd anti-apoptotic role of Hsp60 depending on apoptoticontext.20 A role of Hsp60 protein in cancer is alsoncertain, as its upregulation has been described in sev-ral cancers, such as colorectal21 or cervical cancer.22 Byontrast, positive Hsp60 expression in esophageal squa-ous cell carcinoma23 as well as in ovarian cancer24

orrelates with good prognosis. In prostate cancer, over-xpression of Hsp60 protein was found in early prostatearcinogenesis.25 In other studies, expression of Hsp60as significantly increased in both early and advancedrostate cancer when compared with non-neoplastic ep-thelium, although no consistent correlation was foundetween levels of Hsp60 expression and phenotypic be-avior of primary prostate cancers.26,27 A recent studyith radical prostatectomies, correlated Hsp60 expres-

ion with Gleason score and biochemical recurrence.28 Inelation to hormone resistance, it has been reported thepregulation of Hsp60 in LNCaP-r cells, a cell line derivedrom LNCaP-FGC that does not proliferate in response tondrogen,27 suggesting a possible role in resistance to an-rogens.

Our aim was to study Hsp60 expression using immu-ohistochemical methods in a large series of prostateancers comprising localized and locally advanced carci-omas. We also analyzed the association of tissue Hsp60xpression with established clinicopathological parame-ers of prostate cancer, as well as the relationship be-ween the expression of Hsp60 and the onset of hor-one-resistant phenotype and survival. Identification ofarkers such Hsp60 in diagnostic biopsy samples could

e useful for the clinical management of prostate cancer.

ATERIAL AND METHODS

atients and Tissue Samplese studied 107 patients with prostate cancer, 31 radical pros-

atectomies, and 76 diagnostic needle biopsies. Data regardingleason scores were available for all patients. Initial serum PSA

evels and T classification were available for 106, lymph nodetatus for 45, and distant metastasis for 50 patients.29 In theroup of patients with locally advanced cancer, 50 were eligibleor androgen ablation therapy (ie, locally advanced carcinomatages T3c-T4, or carcinoma with evidence of metastatic spread�/M�). Informed consent was required from patients accord-

ng to the policies of our local Ethical Committee. Exclusionriteria were evidence of concomitant inflammatory disease ofhe prostate, urolithiasis, or the presence of permanent vesicalatheterization. Also, patients who either rejected biopsy pro-edures or fail to give informed consent were not eligible.istopathological diagnosis was established on echography-di-

ected transrectal biopsy samples. Patients were followed every m

017.e2

months in a specialized prostate cancer unit at our hospital.ndrogen-dependent patients were defined on the basis of

ymptoms relief and a decrease of at least 50% of initial serumSA level. Conversion to androgen-independent status wasefined whenever an increase in serum PSA determination withestosterone levels below 50 ng/mL (castration levels), an in-rease of tumor size, or new metastases were detected. Tissueiopsy were fixed in 10% buffered formalin and embedded inaraffin blocks, following standard procedures at the Depart-ent of Pathology. In selected cases, 1 unfixed tumor tissue

ylinder was snap-frozen in liquid nitrogen for western blotnalysis.

mmunohistochemistryissue sections (5 �m) from paraffin blocks were dewaxed and

ehydrated. Sections were immersed in 3% H2O2 aqueous so-ution for 30 minutes to quench endogenous peroxidase, andhen covered with 10% normal swine serum (Vector Laborato-ies, Burlingame, CA) in Tris-buffered saline to block nonspe-ific binding sites. A polyclonal goat anti-Hsp60 antibody wasvailable from Santa Cruz (Santa Cruz, CA). Sections werencubated overnight with a 1:500 dilution of primary anti-sp60 antibody. After several washes in Tris buffer, biotinyl-

ted antigoat immunoglobulin was applied for 15 minutes andashed twice in Tris buffer before application of horseradisheroxidase-labeled streptavidin-biotin complex (LSAB�, Dako,lostrup, Denmark) for 15 minutes at room temperature. Vi-

ualization of the immunoreaction was then achieved using,3=-diaminobenzidine (Dako) as the chromogenic substrate.he slides were then counterstained with Harris hematoxylinnd mounted in DPX. Sections where primary antibody wasmitted were used as negative controls. Immunostaining wasvaluated independently by 2 observers on at least 10 micro-copic fields at �200 magnification and unequivocally stainedells were counted. The immunohistochemical expression wascored as follows: high, �50%; and low/moderate, �50% of thearcinoma cells immunostained.

estern Blottingestern blotting was performed essentially as previously de-

cribed.30 Briefly, 7 selected cases with available frozen materialtumor cell content over 75%), were disrupted and homoge-ized in RIPA buffer (1% Nonidet P-40, 0.5% sodium deoxi-holate, 0.1% sodium dodecyl sulfate, in phosphate-bufferedaline [PBS]) containing protease inhibitors (1:100 dilution ofrotease Inhibitors Cocktail (Sigma, St. Louis, MO) and sub-

ected to sodium dodecyl sulfate–polyacrylamide gel electro-horesis (SDS-PAGE). Dilutions of primary antibodies usedere goat anti-Hsp60 (Santa Cruz), 1:3000; and mouse anti–-actin (Sigma), 1:5000. Secondary peroxidase-labeled anti-odies used were rabbit anti-goat (1:40 000; Sigma) or goatnti-mouse (1:20 000; GE Healthcare Little Chalfont, UK).canning densitometry was performed with QuantiScan soft-are (BioSoft, Cambridge, UK). Arbitrary densitometric unitsf Hsp60 protein were corrected for those of �-actin.

tatistical Analysisssociation between Hsp60 and clinical parameters was ana-

yzed by Fisher’s exact test. Survival and hormone-refractoryisease-free survival curves were estimated by the actuarial

ethod of Kaplan and Meier, and the differences were analyzed

UROLOGY 76 (4), 2010

bsmC

R

IETnemhcpWmtHlplt

Wscm2Gt

AHCWi

FiHwiesg

Fcaew7

U

y the log-rank test of Mantel and Cox. All tests were two-ided, and P � .05 was considered significant. Calculations wereade with the aid of SPSS 15.0 statistical package (SPSS, Inc.,

igure 1. Immunohistochemical Hsp60 protein expressionn benign prostatic hyperplasia and carcinoma tissues. (A)yperplastic glands. Hsp60 immunoreactivity is observedithin the layer of basal cells. The intensity of immunostain-

ng varies between glands. (B) Prostatic carcinoma of mod-rate Gleason score. (C) Prostatic carcinoma of high Glea-on score mixed with unstained non-neoplastic prostatelands.

hicago, IL). a

ROLOGY 76 (4), 2010

ESULTS

mmunohistochemicalxpression of Hsp60 in Prostate Cancero examine the expression of Hsp60, we realized an immu-ohistochemical analysis of 107 formalin-fixed and paraffin-mbedded localized and locally advanced prostate carcino-as biopsy samples. Hsp60 immunostaining was quiteomogeneous, cytoplasmic, and coarsely granular and wasonfined almost exclusively to the epithelial cells of neo-lastic tissues. The stroma did not show immunoreactivity.ell-differentiated, low and moderate Gleason score tu-ors showed low levels of Hsp60 immunostaining, in con-

rast to high Gleason score tumors that exhibited elevatedsp60 expression (Fig. 1). High-grade prostatic intraepithe-

ial neoplasia variably expressed Hsp60 protein and hyper-lastic tissue specimens were occasionally stained but theabeling was restricted to basal cells. In normal prostateissues we did not detect appreciable staining for Hsp60.

Expression of Hsp60 in tumors was also investigated byestern blot in 7 selected cases of different Gleason

cores with available frozen material (Fig. 2). A bandorresponding to 60 kDa was detected. Average densito-etric values of normalized Hsp60 expression showed

.3-fold increase in Gleason scores 7-9 compared withleason scores 3-6, a finding that was in accordance with

he immunohistochemical observations.

ssociation ofsp60 Expression Withlinicopathological Variables of Prostate Carcinomae performed a statistical analysis of immunohistochem-

cal Hsp60 protein expression that showed a significant

igure 2. (A) Western blot analysis of Hsp60 in 7 prostateancers of increasing Gleason scores. (B) Densitometricnalysis of Hsp60 expression levels normalized to �-actinxpression levels. Average normalized values of Hsp60ere calculated for Gleason scores 3-6 and Gleason scores-9 tumors. Data are presented as mean � SEM.

ssociation with Gleason score (P � .001). Most (89.8%)

1017.e3

lemtt.tssrbgat

paae.

AWTcrHaaTefi3tima

dpHacH1m(6

wpweH5l4MMb

CTprmitaceala

1

ow Gleason-score tumors3-6 showed low levels of Hsp60xpression. Moreover, most high Hsp60-expressing tu-ors (58.3%) were of Gleason scores 7-10. Immunohis-

ochemical expression of Hsp60 was also associated withhe presence of regional lymph node metastasis (P �005) and initial serum PSA levels (P � .02). Neitherumor stage nor distant metastases were significantly as-ociated with immunohistochemical Hsp60 overexpres-ion. All these results are summarized in Table 1. We alsoealized a univariate analysis to assess the relationshipetween Hsp60 expression and PSA doubling time in theroup of patients with locally advanced cancer. Thenalysis did not show a significant association betweenhese variables (data not shown).

We also examined Hsp60 expression in the group ofatients with localized cancer versus those with locallydvanced cancer. Expression of Hsp60 was low in 93.8%nd high in 6.2% of localized cancers. By contrast, Hsp60xpression was low in 63.3% and high in 36.7% (P �02) of locally advanced cancers.

ssociation of Hsp60 Expressionith Hormone Resistance and Survivalo assess the relationship between Hsp60 immunohisto-hemical expression and the appearance of hormoneesistance and survival, we performed the analysis ofsp60 expression in the group of patients with locally

dvanced disease. Fifty patients were treated by completendrogen blockade and followed-up every 3 months.heir diagnostic biopsy samples were analyzed for thexpression of Hsp60 by immunohistochemistry. We per-ormed a Kaplan–Meier analysis classifying these tumorsnto 2 groups, high (n � 17), and low/moderate (n �3), according to the expression of Hsp60, and evaluatinghe time in months until the appearance of hormone-ndependent phenotype as the time variable. After aean follow-up of 29 months (range 2-53 months) after

Table 1. Association of Hsp60 immunohistochemical exprcarcinomas

Variable No. Patien

T classification, n (%)T1-T2 90T3-T4 16

Lymph node metastasis, n (%)Yes 25No 20

Distant metastasis, n (%)Yes 32No 18

Gleason score, n (%)3-6 597-10 48

Initial serum PSA level (ng/mL), n (%)0-10 35�10 71

* P value from Fisher’s exact test.

ndrogen ablation, a hormone-independent status was f

017.e4

emonstrated in 22 of 33 patients with low Hsp60 ex-ression (66.6%) and in 16 of 17 patients with highsp60 expression (94.1%). The mean time for achievinghormone-independent status was 35.3 months (95%

onfidence interval [CI], 29.4-41.3) in patients with lowsp60-expressing tumors and 23.29 months (95% CI,

6.79-29.78) in patients with high Hsp60-expressing tu-ors. Kaplan–Meier curves were significantly different

P � .02 from Mantel and Cox, log-rank test; �2 �.485; 1 df) between both groups (Fig. 3A).Next, we performed another Kaplan–Meier analysis

ith the same groups of patients evaluating Hsp60 ex-ression in relation to survival. Cancer-specific survivalas demonstrated in 22 of 33 patients with low Hsp60xpression (66.6%) and in 4 of 17 patients with highsp60 expression (23.5%). Mean time of survival was

3.47 months (95% CI, 44.59-62.35) in patients withow Hsp60 expression and 41 months (95% CI, 32.6-9.48) in those with high Hsp60 expression. Kaplan–eier curves were significantly different (P � .05 fromantel and Cox, log-rank test; �2 � 4.17; 1 df) between

oth groups (Fig. 3B).

OMMENThe molecular chaperone Hsp60 is a protein involved inrotein folding, assembly, and transport. It also plays aole in the apoptotic process, although its function re-ains unclear.20 In relation to cancer, Hsp60 has been

nvestigated in diverse malignancies,21-26 but the rela-ionship between Hsp60 expression with tumor prognosisnd clinical outcome is also controversial.22,23 In prostateancer tissues, elevated Hsp60 expression was found inarly prostate carcinogenesis25 and in both early anddvanced cancer with respect to normal prostate epithe-ium,26,27 although these studies did not find any associ-tion with Gleason score. In contrast, Glaessgen et al.28

on with selected clinicopathological variables of prostatic

Hsp60 Expression P Value*

Low/moderate High

65 (72.2) 25 (27.8) .0878 (50.0) 8 (50.0)

10 (40.0) 15 (60.0) .00317 (85.0) 3 (15.0)

17 (53.1) 15 (46.9) .76811 (61.1) 7 (38.9)

53 (89.8) 6 (10.2) �.000120 (41.7) 28 (58.3)

30 (85.7) 5 (14.3) .0143 (60.6) 28 (39.4)

essi

ts

ound a correlation between Hsp60 expression and Glea-

UROLOGY 76 (4), 2010

smtwsc

Hcstanstf

ew

lgihlttaLftppdshattdldpw

CHpMhevaededws

AtICsEFIS

R

FdaeoiP

U

on score in their series of radical prostatectomies. Inultivariate analysis, Hsp60 was an independent predic-

or of biochemical recurrence after radical prostatectomyhen extraprostatic extension, positive surgical margins,

eminal vesicles invasion, and Gleason score were in-luded in the analysis.

Concordantly with these authors, we found elevatedsp60 immunohistochemical expression in prostate can-

ers in comparison with normal tissue. We also found atatistically significant relationship between immunohis-ochemical Hsp60 expression and Gleason score, as wells with other prognostic tumor variables, such as lymphode metastasis and initial serum PSA levels. Also, in oureries of prostate cancer patients, normal and hyperplas-ic tissues were largely unstained, with the exception ofocally stained basal cells.

To further examine the possible implications of Hsp60xpression in the hormone-responsive status of patients,

igure 3. (A) Kaplan–Meier analysis of hormone-refractory,isease-free survival in patients showing either low/moder-te (solid line, *n � 33) or high (dotted line, **n � 17)xpression of Hsp60 (P � .02). (B) Kaplan–Meier analysisf cancer-specific survival in the same groups of patients as

n A (P � .05). Tick marks represent censored patients.value from log-rank test of Mantel and Cox.

e performed a Kaplan–Meier analysis in the group of

ROLOGY 76 (4), 2010

ocally advanced carcinomas treated by complete andro-en blockade. We found a significant association betweenmmunohistochemical Hsp60 expression and the onset oformone refractory disease, i.e, those patients with high

evels of Hsp60 expression showed a significantly shorterime for the develop of hormone-refractory disease thanhose with low Hsp60 levels. Interestingly, Johansson etl.27 reported the upregulation of Hsp60 protein in theNCaP-r cell line, an androgen-resistant cell line derived

rom LNCaP-FGC. These results agree with our observa-ions in prostate cancer biopsy samples, suggesting aossible role of Hsp60 in hormone resistance. Moreover,atients with high Hsp60 expression had shorter times toisease-related death, whereas patients with low expres-ion levels had longer survival. Several in vitro studiesave pointed to a cytoprotective role for Hsp60,18,19

lthough others are in line with a proapoptotic func-ion.16 In prostate cancer cell lines, specifically, func-ional data exist supporting a dual pro- or anti-death roleepending on the apoptotic context.20 In the case ofocally advanced prostate cancers treated by androgeneprivation, Hsp60 could exert an antiapoptotic role,rovided that therapeutic failure occurs earlier in patientsith high Hsp60 expression.

ONCLUSIONSsp60 protein is overexpressed in prostate cancers com-

ared with normal prostate epithelium and hyperplasia.oreover, high Gleason score tumors had significantly

igher levels of Hsp60 protein than tumors of low-mod-rate Gleason score. Finally, in the group of locally ad-anced cancers treated with androgen ablation, we foundsignificant relationship between high levels of Hsp60

xpression and an earlier onset of hormone-refractoryisease and reduced cancer-specific survival. Thus Hsp60xpression is highly associated with androgen indepen-ence. This protein could be useful to identify patientsith prostate cancer more likely to have disease progres-

ion and to develop a hormone-resistant phenotype.

cknowledgments. This work was supported by grants fromhe Spanish Ministerio de Salud, Instituto de Salud CarlosII/FIS (FIS 06/757), ISCIII-RETIC-RD06/0020-FEDER, andonsejería de Salud, Junta de Andalucía (CS 06/89). C.C. was

upported by a predoctoral grant from the Spanish Ministerio deducación y Ciencia (FPI: BES-2006-12419) co-financed byondo Social Europeo. C.S. was supported by a contract fromnstituto de Salud Carlos III/FIS and Fundación Progreso yalud, Consejería de Salud Junta de Andalucía.

eferences1. Boyle P, Severi G, Giles GG. The epidemiology of prostate cancer.

Urol Clin North Am. 2003;30:209-217.2. Saric T, Brkanac Z, Troyer DA, et al. Genetic pattern of prostate

cancer progression. Int J Cancer. 1999;81:219-224.3. Tsihlias J, Kapusta LR, DeBoer G, et al. Loss of cyclin-dependent

kinase inhibitor p27Kip1 is a novel prognostic factor in localized

human prostate adenocarcinoma. Cancer Res. 1998;58:542-548.

1017.e5

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

1

4. Gu K, Mes-Masson AM, Gauthier J, et al. Analysis of the p16tumor suppressor gene in early-stage prostate cancer. Mol Carcinog.1998;21:164-170.

5. Taplin ME, Bubley GJ, Shuster TD, et al. Mutation of the andro-gen-receptor gene in metastatic androgen-independent prostatecancer. N Engl J Med. 1995;332:1393-1398.

6. Chen CD, Welsbie DS, Tran C, et al. Molecular determinants ofresistance to antiandrogen therapy. Nat Med. 2004;10:33-39.

7. Hartl FU, Hayer-Hartl M. Molecular chaperones in the cytosol:from nascent chain to folded protein. Science. 2002;295:1852-1858.

8. Beere HM. Death versus survival: functional interaction betweenthe apoptotic and stress-inducible heat shock protein pathways.J Clin Invest. 2005;115:2633-2639.

9. Christians ES, Zhou Q, Renard J, et al. Heat shock proteins inmammalian development. Semin Cell Dev Biol. 2003;14:283-290.

0. Soltys BJ, Gupta RS. Mitochondrial proteins at unexpected cellularlocations: export of proteins from mitochondria from an evolution-ary perspective. Int Rev Cytol. 2000;194:133-196.

1. Deocaris CC, Kaul SC, Wadhwa R. On the brotherhood of themitochondrial chaperones mortalin and heat shock protein 60. CellStress Chaperones. 2006;11:116-128.

2. Oba M, Yano S, Shuto T, et al. IFN-gamma down-regulates Hsp27and enhances hyperthermia-induced tumor cell death in vitro andtumor suppression in vivo. Int J Oncol. 2008;32:1317-1324.

3. Mosser DD, Caron AW, Bourget L, et al. The chaperone functionof hsp70 is required for protection against stress-induced apoptosis.Mol Cell Biol. 2000;20:7146-7159.

4. Chakraborty A, Koldobskiy MA, Sixt KM, et al. HSP90 regulatescell survival via inositol hexakisphosphate kinase-2. Proc Natl AcadSci U S A. 2008;105:1134-1139.

5. Samali A, Cai J, Zhivotovsky B, et al. Presence of a pre-apoptoticcomplex of pro-caspase-3, Hsp60 and Hsp10 in the mitochondrialfraction of Jurkat cells. EMBO J. 1999;18:2040-2048.

6. Xanthoudakis S, Roy S, Rasper D, et al. Hsp60 accelerates thematuration of pro-caspase-3 by upstream activator proteases duringapoptosis. EMBO J. 1999;18:2049-2056.

7. Kirchhoff SR, Gupta S, Knowlton AA. Cytosolic heat shock protein60, apoptosis, and myocardial injury. Circulation. 2002;105:2899-2904.

8. Lin KM, Lin B, Lian IY, et al. Combined and individual mitochon-drial HSP60 and HSP10 expression in cardiac myocytes protects

mitochondrial function and prevents apoptotic cell deaths induced

017.e6

by simulated ischemia-reoxygenation. Circulation. 2001;103:1787-1792.

9. Ghosh JC, Dohi T, Kang BH, et al. Hsp60 regulation of tumor cellapoptosis. J Biol Chem. 2008;283:5188-5194.

0. Chandra D, Choy G, Tang DG. Cytosolic accumulation of HSP60during apoptosis with or without apparent mitochondrial release.J Biol Chem. 2007;282:31289-31301.

1. Cappello F, Bellafiore M, Palma A, et al. 60kDa chaperonin(HSP60) is over-expressed during colorectal carcinogenesis. EurJ Histochem. 2003;47:105-110.

2. Castle P, Ashfaq R, Ansari F, et al. Immunohistochemical evalu-ation of heat shock proteins in normal and preinvasive lesions ofthe cervix. Cancer Lett. 2005;229:245-252.

3. Faried A, Sohda M, Nakajima M, et al. Expression of heat-shockprotein Hsp60 correlated with the apoptotic index and patientprognosis in human oesophageal squamous cell carcinoma. Eur JCancer. 2004;40:2804-2811.

4. Schneider J, Jimenez E, Marenbach K, et al. Immunohistochemicaldetection of HSP60-expression in human ovarian cancer. Correla-tion with survival in a series of 247 patients. Anticancer Res.1999;19:2141-2146.

5. Cappello F, Rappa F, David S, et al. Immunohistochemical evalu-ation of PCNA, p53, HSP60, HSP10 and MUC-2 presence andexpression in prostate carcinogenesis. Anticancer Res. 2003;23:1325-1331.

6. Cornford PA, Dodson AR, Parsons KF, et al. Heat shock proteinexpression independently predicts clinical outcome in prostatecancer. Cancer Res. 2000;60:7099-7105.

7. Johansson B, Pourian MR, Chuan YC, et al. Proteomic comparisonof prostate cancer cell lines LNCaP-FGC and LNCaP-r revealsheatshock protein 60 as a marker for prostate malignancy. Prostate.2006;66:1235-1244.

8. Glaessgen A, Jonmarker S, Lindberg A, et al. Heat shock proteins27, 60 and 70 as prognostic markers of prostate cancer. APMIS.2008;116:888-895.

9. Sobin LH, Wittekind Ch, eds. International Union Against Cancer(UICC): TNM Classification of Malignant Tumours, 6th ed. NewYork: Wiley-Liss; 2002.

0. Castilla C, Congregado B, Chinchón D, et al. Bcl-xL is overex-pressed in high-grade prostate cancer and promotes survival ofLNCaP cells via interaction with pro-apoptotic bak. Endocrinology.

2006;147:4960-4967.

UROLOGY 76 (4), 2010