Urologic Oncology: Seminars and Original Investigations 32 (2014) 51.e9–51.e19

Original article

The prognostic value of ribonucleotide reductase small subunit M2 in predicting recurrence for prostate cancers Yasheng Huang, M.D.a,b,1, Xiyong Liu, M.D., Ph.D.b,1, Yuan-Hung Wang, Ph.D.c,d, Shauh-Der Yeh, M.D., Ph.D.e, Chi-Long Chen, M.D.f,g, Rebecca A. Nelson, Ph.D.h, Peiguo Chu, M.D., Ph.D.i, Timothy Wilson, M.D.j, Yun Yen, M.D., Ph.D.k,l,* a Department of Urology, Hangzhou Traditional Chinese Medical Hospital, Hangzhou, Zhejiang, China Department of Molecular Pharmacology, City of Hope National Medical Center and Beckman Research Institute, Duarte, CA c Division of Urology, Department of Surgery, Shuang Ho Hospital, Taipei Medical University, Taipei, Taiwan (R.O.C) d Graduate Institute of Clinical Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan (R.O.C) e Department of Urology, Taipei Medical University; Taipei, Taiwan (R.O.C.) f Department of Pathology, Wan Fang Hospital, Taipei Medical University, Taiwan (R.O.C.) g Department Center of Excellence for Cancer Research, Taipei Medical University, Taipei, Taiwan (R.O.C.) h Department of Information Science, City of Hope National Medical Center and Beckman Research Institute, Duarte, CA i Department of Pathology, City of Hope National Medical Center and Beckman Research Institute, Duarte, CA j Department of Surgery PS-Urology & Urologic Oncology, City of Hope National Medical Center and Beckman Research Institute, Duarte, CA k Department of Molecular Pharmacology, City of Hope National Medical Center and Beckman Research Center, Duarte, CA l The Ph.D. Program for Cancer Biology and Drug Discovery, Taipei Medical University, Taipei, Taiwan (R.O.C) b

Received 29 May 2013; received in revised form 25 July 2013; accepted 5 August 2013

Abstract Purpose: To investigate the prognostic significance of ribonucleotide reductase small subunit M2 (RRM2) in low- and intermediate-risk prostate cancer (PCa). Materials and methods: A retrospective outcome study was conducted on 164 eligible PCa samples from the City of Hope (n ¼ 90) and the Taipei Medical University (n ¼ 74). The RRM2 protein levels were detected by immunohistochemistry. Biochemical recurrence was assessed using Kaplan-Meier and Cox proportional hazard analyses. Cell invasion assays, Ras/Raf, and matrix metallopeptidase 9 activities were determined to evaluate the role of RRM2 on invasiveness of PCa. Results: Expression of RRM2 was significantly increased in patients with higher Gleason score, who had advanced T stage, and who were margin/capsule positive (P o 0.05). Analysis revealed that the expression of RRM2 positively associated with biochemical recurrence of PCa in the City of Hope set (hazard ratio ¼ 5.26; 95% CI 1.50–24.71) and the Taipei Medical University set (hazard ratio ¼ 2.55; 95% CI 1.30–9.22). In stratification analysis, RRM2 was significantly correlated with poor outcome in patients with lower-risk PCa, including those with Gleason score 4 to 7, margin, capsule, and stage T1-T2. In patients with Gleason score 4 to 7, the risk of recurrence was proportional to RRM2 protein levels. The prognostic performance of RRM2 was superior to that of pathoclinical factors, including margin/ capsule status and T stage. An in vitro study demonstrated that RRM2 could promote tumor invasion activities in PCa cell lines. Suppression of RRM2 reduced the Ras/Raf and matrix metallopeptidase 9 activities. Conclusion: RRM2 plays a critical role in proliferation and invasion of PCa. Adding RRM2 as a biomarker in clinical assessments may increase model precision in predicting recurrence in patients with low-risk PCa. r 2014 Elsevier Inc. All rights reserved. Keywords: Ribonucleotide reductase M2; Prostate cancer; Prognostic biomarker; Gleason score; Biochemical recurrence

1. Introduction Corresponding author. Tel.: þ1-626-256-4673, ext: 65707; fax: þ1626-471-3607. E-mail address: [email protected] (Y. Yen). 1 Yasheng Huang and Xiyong Liu are co-first authors and contributed equally to this study. *

1078-1439/$ – see front matter r 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.urolonc.2013.08.002

Prostate cancer (PCa) is the most common malignancy and the second leading cause of mortality in Western countries. According to estimation, it accounted for 29% (238,590 of 854,790) of total new cases and 10% (29,720 of

51.e10

Y. Huang et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 51.e9–51.e19

306,920) of total cancer deaths in men in 2013 [1]. Since the emergence and dissemination of prostate-specific antigen (PSA) screening, the mortality because of PCa has decreased by nearly one-third in the United States [2]. However, the clinical behavior of PCa varies dramatically, from indolent to aggressive [3], leading to a dilemma when determining treatment options. Currently, increasingly more PCas are being diagnosed in population-based mass screening [4]. In the European Randomized Study of Screening for Prostate Cancer, more than 600 men with clinically and pathologically defined low-risk PCa features were observed without primary treatment for 10 years. The overall survival was 70%, although no one died of PCa [5]. This implies that a subset of men diagnosed with PCa do not require any active, invasive treatment. However, approximately one-third of patients who received local therapies eventually underwent biochemical recurrence with elevated PSA [6]. This was probably owing to undetected dissemination of cancer cells or insufficient local therapy. A critical issue for PCa is to improve identification of the risk for tumor recurrence, so that therapy can be targeted to patients who would benefit, simultaneously avoiding overtreatment of patients at low risk of recurrence. PSA is the main tool for risk stratification and recurrence monitoring in PCa, with Gleason score also widely accepted in clinical practice [7]. According to the 2012 National Comprehensive Cancer Network guideline for PCa, only patients with high-risk PCa (Gleason score 8–10, stage T3-T4, and metastasis) need more aggressive treatment. It remains difficult to decide whether a patient with a relatively low Gleason score should be treated. Lack of prognostic and therapeutic biomarkers is currently a barrier for deciding on therapeutic regimens and reducing mortality of patients with low-risk (Gleason score r6) or intermediate-risk (Gleason score ¼ 7) PCa. Thus, there is a real need for predictive biomarkers to identify higher risk of recurrence among patients with relatively low Gleason score. Ribonucleotide reductase (RR) is an essential enzyme for the de novo production of precursors of 2′-deoxyribonucleotide 5′-triphosphates, which are the basic building blocks for DNA synthesis. RR catalyzes the conversion of ribonucleoside diphosphates to the corresponding 2′-deoxyribonucleotide diphosphates in a rate-limiting manner and plays important roles in DNA replication and repair. In mammals, it is a heterodimeric tetramer consisting of 2 large subunits (RRM1) and 2 small subunits (RRM2 and RRM2B) [8]. RRM1 contains an active site and an allosteric site, which control enzyme activity and substrate specificity, respectively. Catalysis requires the binuclear iron center and tyrosyl free radical, which are located in RRM2 [8]. RRM2B, also called p53R2, is the p53-inducible small subunit that substitutes for RRM2 and forms the RR holoenzyme responsible for DNA repair [9]. Previous studies have reported the role of RRM2 as a prognostic or predictive biomarker in several types of malignancies. In pancreatic ductal adenocarcinoma, lower RRM2 messenger RNA level was associated with better overall and disease-free survival in gemcitabine-treated patients [10,11]. In some cancer types, a high level of RRM2 messenger RNA

correlates with chemoresistance [7], cellular invasiveness [12], and poor outcome, suggesting that RRM2 contributes to malignant progression and is a potential therapeutic target. In this study, we investigated the effect of expression levels of RRM2 and RRM2B on the aggressive potential of PC-3 and LNCaP PCa cells. Moreover, we investigated the role of RRM2 protein levels in the clinical outcome of patients with PCa using 90 PCa patient samples from the City of Hope National Medical Center (COH) and 78 PCa patient samples from the Taipei Medical University (TMU). Our findings suggested that RRM2 played a critical role in proliferation and invasion of PCa. It may be a potential prognostic biomarker combined with PSA and Gleason score for intermediate- and low-risk PCas. 2. Material and methods 2.1. Patients and study design After obtaining approval from Institutional Review Boards (IRB COH #06206, TMU #TMU-JIRB 98-09-A), eligible participants with PCa were enrolled and PCa databases were built based on medical record review and follow-up data. Patients were selected according to the following criteria. Inclusion criteria: (1) PCa with pathological diagnosis; (2) diagnosis during 1987 to 2004 (COH) and 1999 to 2011 (TMU); (3) informed consent obtained; and (4) receipt of at least one follow-up. Exclusion criteria: (1) multiple cancers; (2) lack of pathology diagnosis; and (3) lost to follow-up within 3 months. Patients who underwent prostatectomy at the COH (n ¼ 90) and the TMU (n ¼ 74) were recruited in this study. Almost all participants from the COH were white, but all cases from the TMU set were Asian. Clinical and pathology information including age, Gleason score, margin and capsule status, and TNM stage was obtained after careful chart review and are displayed in Supplementary Table S1. Patients from the TMU set (70.8 ⫾ 7.7) were older than the COH set (65.5 ⫾ 8.5). On the contrary, there were more early T stage in the TMU set (Pearson test, P ¼ 0.004). The inconsistence of average age and stage distribution might be caused by differences in race and social-economic background between the 2 sets. All patients were periodically followed up, with the follow-up period being calculated from the date of surgery through date of last follow-up. Follow-up data including biochemical recurrence event (PSA 40.4 ng/ml) [13], biochemical recurrence time, and status during last follow-up were collected into the database. Duration of disease-free survival was defined as the time from initial surgery to the time of biochemical recurrence. Only death because of PCa was considered as cancer-specific death. 2.2. Immunohistochemistry The human prostate tissue sections were first deparaffinized and rehydrated in graded alcohol. After deparaffinization, the

Y. Huang et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 51.e9–51.e19

endogenous peroxidase activity was blocked by using 3% H2O2, 0.1 mM EDTA buffer (pH 8.0) at 1031C for 30 min in a pressure cooker (Biocare Medical, Concord, CA). Sections were incubated with normal goat serum (20 min), and then exposed to primary antibody (20 min) at room temperature. After hydrogen peroxide treatment (7 min), the array slides were incubated (30 min) with labeled polymer horseradish peroxidase corresponding antibodies (a goat polyclonal antibody against human RRM2, Santa Cruz Biotechnology, Santa Cruz, CA). Samples were treated with 3,3′-diaminobenzidine (0.05 g 3,3′-diaminobenzidine and 100 ml 30% H2O2 in 10 ml phosphate-buffered saline) for 10 min. Phosphate-buffered saline was used as a negative control. Firstly, we optimized semiquantitative immunohistochemical assays for detecting RRM2 in formalin-fixed, paraffin-embedded human prostate tissue array sections, and then the same assay was employed to detect the RRM2 expression level in the collected tissue samples. Two independent observers viewed the slides in a double-blind manner. The subcellular localization, staining intensity, and percentage of stained cells were evaluated for each slide after joint review by the 2 readers. We observed RRM2 and RRM2B mainly in the cytoplasm. RRM2 and RRM2B expression were quantified by a visual grading system that took into consideration the percentage of positive cells and staining intensity (0, negative; 1, weak; 2, moderate; and 3, strong); For further analysis, RRM2 and RRM2B moderate positive and strong positive were attributed as RRM2 and RRM2B high and others as RRM2 and RRM2B low. 2.3. Cell culture expression plasmid transfection and small interfering knockdown Human PCa cell lines PC-3 and LNCaP were obtained from American Type Culture Collection and cultured in RPMI-1640 medium (Sigma Aldrich, St Louis, MO), supplemented with 10% fetal bovine serum (MP Biomedicals, Costa Mesa, CA) and 1% penicillin and streptomycin in a humidified atmosphere containing 5% CO2 at 371C. The strands of sense complementary DNA of RRM2 and RRM2B were inserted into pcDNA 3.1 expression vectors to construct expression plasmids. The expression plasmids, including control vector pcDNA 3.1, were transfected into PC-3 and LNCaP cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Stable transfectants were selected with 400 μg/ml G418 for 4 weeks. The RRM2 and RRM2B protein levels were determined by Western blot lysates of the stable clones: PC-3V, PC-3RRM2, PC-3RRM2B, LNCaP-V, LNCaP-RRM2, and LNCaP-RRM2B. RRM2, RRM2B, and scramble small interfering (siRNA) were purchased from Santa Cruz Biotechnology Inc. PC-3 and LNCaP cells were seeded at 2  105 cells per well in 6-well plates filled with 2 ml antibiotic-free normal growth medium, supplemented with fetal bovine serum and then incubated at 371C in CO2 incubator for 24 hours. A total of

51.e11

4 ml of 10 nmol/ml RRM2, RRM2B, or scramble siRNA was transfected into PC-3 or LNCaP cells by using a transfection RNAiMAX reagent (Invitrogen, Carlsbad, CA). Cells were incubated in the transfection medium for 6 hours and then placed in normal cell culture medium. The inhibition of RRM2 and RRM2B was measured by using Western blot analysis. 2.4. Ras activity assay Ras-activation assay kit was obtained from EMD Millipore (a division of Merck KGaA, Darmstadt, Germany). In vitro Ras activity was stimulated by 24-hour incubation of 50% to 60% confluent cultures of PC-3 or LNCaP cells in serum-free and antibiotic-free medium. Medium was then removed and cells were rinsed with ice-cold phosphatebuffered saline before the addition of ice-cold lysis buffer (1 ml). Lysed cells were harvested with a rubber policeman and the lysates were cleared of insoluble cell debris by microcentrifugation (5 min, 14,000g, and 41C). Aliquots (300 ml each) of the extracts were added to 3 tubes: (1) positive control, containing 6 ml 0.5 M EDTA and 3 ml 100X GTPγS; (2) negative control, containing 6 ml 0.5 M EDTA and 3 ml 100X GDP; and (3) extract alone. Positive and negative controls (tubes 1 and 2) were first incubated at 301C with agitation for 30 minutes, before loading was stopped by placing the tubes on ice and adding 19.5 ml of 1 M MgCl2. Ras assay reagent (7 ml) was added to each tube. Tubes 1, 2, and 3 reaction mixtures were incubated for 45 minutes at 41C with gentle agitation. The agarose beads were pelleted by brief centrifugation (10 s, 14,000g, and 41C). The supernatant was removed and discarded, and the beads were washed 3 times with Mg2+ Lysis/wash buffer. The agarose beads were resuspended in 40 ml of 2 laemmli buffer and boiled for 5 minutes. The supernatant and agarose pellets were mixed, and 20 ml of mixture was loaded per lane on 20% polyacrylamide gels. 2.5. Western blot analysis Western blot analysis was performed as described previously [14]. Anti-RRM2 antibody (0.2 mg/ml) or goat polyclonal anti-β-actin antibody (Santa Cruz Biotechnology, 0.2 mg/ml) was used as primary antibodies. AntiRRM2B antibody and antitubulin antibody were obtained from Santa Cruz Biotechnology (0.2 mg/ml). 2.6. Gelatin zymography assay Matrix metallopeptidase 9 (MMP-9) activity in the cell conditioned medium was analyzed using standard gelatin zymography with reagents purchased from Invitrogen. Cells were cultured in serum-free RPMI-1640 medium for 24 hours and conditioned medium was collected. Protein in the medium was concentrated with an AmiconUltracel 30 KDa filter (Millipore, Billerica, MA). Equal amounts of protein

51.e12

Y. Huang et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 51.e9–51.e19

(10 mg per sample) were mixed with 26 Novex Tris-Glycine sodium dodecyl sulfate sample buffer, and fractionated on a 10% gelatin gel under nonreducing conditions. The gel was then incubated at 371C in renaturing buffer for 60 minutes and in developing buffer for 16 hours. Finally, the gel was stained in SimplyBlue SafeStain, and bands representing the gelatinase activity of MMP-9 were quantified. 2.7. Invasion assay A 24-well invasion chamber was purchased from BD Company. For the invasion assay, 2.5  104 PC-3V, PC-3RRM2, PC-3RRM2B or LNCaP-V, LNCaP-RRM2, and LNCaP-RRM2B cells were seeded on the Matrigel insert of the 24-well chamber. After incubation for 72 hours in 5% CO2 at 371C, the tops of the Matrigel inserts were wiped with a cotton-tipped swab to remove cells that had not migrated through the membrane. The cells on the lower surface of the membrane were stained with 0.5% Coomassie blue (dissolved in 50% ethanol) and counted. Each experiment was performed 3 times. 2.8. Data management and statistical analysis

Table 1 High expression of RRM2/RRM2B and pathoclinical features of patients with prostate cancer in the COH set Case

RRM2 higha n (%)

Age o60 60–69 Z70

RRM2B highb P valuec

n (%)

P valuec

0.843

3 (14.3) 18 (42.9) 9 (33.3)

0.076

0.156

21 42 27

12 (57.1) 22 (52.4) 15 (57.7)

Gleason grade 4–6 24 7 23 8–10 21

11 (45.8) 15 (65.2) 18 (85.7)

0.016

8 (26.7) 9 (27.3) 13 (48.2)

Stage T1-2 T3

58 30

27 (46.6) 22 (73.3)

0.017

18 (31.0) 11 (36.7)

0.594

Margin  þ

55 34

26 (47.4) 24 (70.6)

0.031

16 (29.1) 14 (41.2)

0.241

Capsule  þ

60 30

28 (46.7) 22 (73.3)

0.016

19 (31.7) 11 (36.7)

0.635

a

RRM2 high: RRM2 (moderate) or RRM2 (strong). RRM2B high: RRM2B (moderate) or RRM2B (strong). c Pearson chi-square test. b

Data were collected using MS-Access and analyzed using the JMP Statistical Discovery Software version 8.0 (SAS Institute, Cary, NC), and GraphPad Prism 5.0 was used to generate figures. Group comparisons for continuous data were performed using t tests for independent means or oneway analyses of variance. For categorical data, we employed chi-square analyses, Fisher exact tests, or binomial tests of proportions. Kaplan-Meier and recurrence analyses were employed to assess biochemical recurrence in this study, with multivariate logistic regression model and Cox hazard proportional model used to adjust for covariate effects on the odds ratio (OR) and hazard proportional ratio (HR). Statistical significance was set at P o 0.05. 3. Results 3.1. Expression of RRM2 is positively associated with progression of PCa RRM2 and RRM2B expression levels were determined by immunohistochemistry (IHC), and the patients were stratified as low (0, or 1) or high (2 or 3) expression subgroups as described in Section 2. The RRM2 and RRM2B expression distributions are listed in Table 1. The expression of RRM2, but not RRM2B, was significantly associated with Gleason score (P ¼ 0.016), tumor stage (P ¼ 0.017), margin involvement (P ¼ 0.031), and capsule integrity (P ¼ 0.016) in the COH set. The Gleason score has been used as a universal prognostic marker to estimate risk of recurrence. Representative sections illustrating negative, moderate, and strong signal for RRM2 or

RRM2B, as well as the corresponding Gleason score, are shown in Fig. 1A. Kaplan-Meier analysis showed that Gleason score was significantly associated with recurrence of PCa in the COH set (Fig. 1B), which is compatible with previous studies. Nonconditional logistic analyses were carried out to determine whether there was a relationship between RRM2 expression and the pathoclinical features of PCa (Table 2). RRM2 high expression was related to an increased OR of higher Gleason score. When the subgroup of Gleason score 4 to 6 was considered as reference, the adjusted ORs for RRM2 were 4.72 (95% CI 1.61–15.08) and 5.46 (95% CI 1.76–18.72) for Gleason 7 and Gleason 8 to 10 subgroups, respectively. High RRM2 was also associated with risk of advanced tumor stage (OR ¼ 3.16; 95% CI 1.25–8.69), margin positive (OR ¼ 2.77; 95% CI 1.12–7.69) and capsule positive (OR ¼ 3.15; 95% CI 1.25–8.55) in the COH set. RRM2B has more than 80% similarity to RRM2, but high RRM2B expression did not relate to Gleason score, tumor stage, margin positivity, and capsule involvement significantly (Supplementary Table S2). The aforementioned findings indicated that RRM2 was significantly and specifically related to aggressiveness of PCa.

3.2. The prognostic significance of RRM2 in biochemical recurrence of PCas Because only 4 subjects died during follow-up, we could not evaluate overall survival in this study. To

Y. Huang et al. / Urologic Oncology: Seminars and Original Investigations 32 (2014) 51.e9–51.e19

H&E

RRM2

51.e13

RRM2B

RRM2 (0)

RRM2B (0)

RRM2 (2)

RRM2B (2)

RRM2 (3)

RRM2B (3)

Gleason 4

Gleason 7

Gleason 9

1.0 0.9

Gleason 4-6 (n=30) Gleason 7 (n=33) Gleason 8-10 (n=27)

COH set

0.8 Recurrence

0.7

Log-rank P

The prognostic value of ribonucleotide reductase small subunit M2 in predicting recurrence for prostate cancers.

To investigate the prognostic significance of ribonucleotide reductase small subunit M2 (RRM2) in low- and intermediate-risk prostate cancer (PCa)...
1MB Sizes 0 Downloads 0 Views