Tumor Biol. (2011) 32:481–487 DOI 10.1007/s13277-010-0141-6

RESEARCH ARTICLE

HuR expression is a marker of poor prognosis in renal cell carcinoma Hanna Ronkainen & Markku H. Vaarala & Pasi Hirvikoski & Ari Ristimäki

Received: 6 October 2010 / Accepted: 1 December 2010 / Published online: 16 December 2010 # International Society of Oncology and BioMarkers (ISOBM) 2010

Abstract The HuR protein is a nucleocytoplasmic protein which plays an important role in the regulation of mRNA stability, and dysregulation of its expression has been linked to carcinogenesis. We studied 152 patients with primary renal cell carcinoma (RCC) who underwent surgery for the removal of kidney tumours between 1990 and 1999. The mean follow-up was 90 months. The expression of HuR and cyclooxygenase-2 (COX-2) was determined by immunohistochemistry using monoclonal antibodies. The immunostaining results were associated with patient age, clinical stage, Fuhrman grade and patient outcome. Cytoplasmic expression of HuR and COX-2 was positive in 37 (25%) and 22 (15%) of the tumours, respectively. The expression of HuR was associated with stage. The expression of COX-2 was associated with stage and nuclear grade. The RCC-specific survival was reduced in patients whose tumours expressed HuR or COX-2. The hazard ratio (HR) of patients with HuR-expressing tumours was 2.18 (95% confidence interval (CI), 1.16–4.09; p= 0.015) and the HR of patients with COX-2-expressing Hanna Ronkainen and Markku H. Vaarala contributed equally. H. Ronkainen : M. H. Vaarala (*) Department of Surgery, Oulu University Hospital, P.O. Box 21, 90029 OYS, Oulu, Finland e-mail: [email protected] P. Hirvikoski : A. Ristimäki Department of Pathology, Oulu University Hospital and University of Oulu, P.O. Box 50, 90029 OYS, Oulu, Finland A. Ristimäki Department of Pathology, HUSLAB and Haartman Institute, and Genome-Scale Biology Research Program, University of Helsinki, P.O. Box 63, 00014 Helsinki, Finland

tumours was 2.29 (95% CI, 1.15–4.54; p=0.018). In the Cox regression analysis the only independent prognostic factor was stage (p<0.001). Treatment of an RCC cell line (769-P) with HuR-targeted small interfering RNA resulted in the reduced expression of HuR and COX-2. We conclude that cytoplasmic HuR expression is associated with reduced RCC-specific survival. The HuR protein regulates the expression of COX-2 in RCC cells, which is one potential mechanism of action for the HuR-associated aggressive behaviour of RCC. Keywords Prognosis . Immunohistochemistry . Surgery . Survival . HuR . COX-2

Introduction Renal cell carcinoma (RCC) is a cancer of increasing incidence and mortality and one of the most lethal urological malignancies with up to 40% of patients eventually dying of cancer progression [1, 2]. Despite several well-characterised prognostic factors for the disease, such as carbonic anhydrase IX and vascular endothelial growth factor (VEGF) [3, 4], the behaviour of RCC remains difficult to predict and new prognostic markers are needed. The HuR protein (also known as ELAVL1, a member of the embryonic lethal, abnormal vision, Drosophila homolog-like (ELAVL) family of RNA-binding proteins) is an omnipresent nucleocytoplasmic protein [5]. It plays an important role in mRNA stability [6] by binding selectively to adenylate/uridylate-rich elements (AREs), stabilising mRNAs [5, 7]. It has been suggested that HuR binds to ARE-containing mRNAs in the nucleus, and that this complex is then transported to the cytoplasm [7]. The

482

Tumor Biol. (2011) 32:481–487

Materials and methods

slides. The tissue sections were deparaffinized in xylene, rehydrated in a descending ethanol series and washed in phosphate-buffered saline (PBS). To enhance immunoreactivity, the sections were boiled in a microwave oven in 0.3% citrate buffer (pH 6.0) for 4×5 min or they were boiled in Tris–EDTA buffer (pH 9.0), for COX-2 and HuR, respectively. They were then washed with PBS for 2× 5 min. Endogenous peroxidase activity was inactivated by 30-min incubation in methanol containing 1.6% hydrogen peroxidise, after which the sections were washed with PBS for 3×5 min. The antibodies used were mouse monoclonal antibody COX-2 (Cayman Chemical, Ann Arbor, MI, USA, dilution 1:200) and mouse monoclonal antibody HuR (19F12, sc-56709, Santa Cruz Biotechnology, Santa Cruz, CA, USA, dilution 1:1,500). The immunohistochemistry with these antibodies was previously described [9, 11]. The bound antibodies were visualised using an EnVision Detection System (K500711; Dako Denmark A/S, Glostrup, Denmark). Diaminobenzidine was used as the chromogen.

Patients

Immunostaining evaluations

This retrospective clinical cohort consisted of 152 patients with 77 (51%) females and 75 (49%) males who underwent surgery for primary renal cell carcinoma between 1990 and 1999 at the Oulu University Hospital. The clinical data and patient follow-up details were collected from patient records. Seven patients (5%) were operated on by resection and 145 (95%) by radical nephrectomy. At the time of diagnosis, the median age of the patients was 63 years old (range, 29–86) and the mean age was 62 (SD±11). The median and mean follow-up times were 90 months (range, 0–209) and 90 months (SD±63), respectively. The followup was completed in all cases. During the follow-up period, 44 (29%) patients died of RCC, 40 (26%) died of other causes and 68 (45%) were still alive. The distribution of clinicopathological parameters of the tumours was previously described [16]. Of the patients, six had lymph node metastases and 18 had distant metastases. The stage of the tumours was assigned using the TNM staging of renal cell carcinoma [17]. Approval of the study was obtained from the local ethical board.

Cytoplasmic HuR immunoreactivity was initially scored according to three cytoplasmic staining intensities: negative (0), weak (1) or strong (2). For further statistical analyses, the negative samples (score 0) were compared to the positive ones (scores 1 and 2). Nuclear immunoreactivity was also assessed and scored as either negative or positive. Cytoplasmic COX-2 immunoreactivity was also assessed by the intensity of staining. The intensity of staining was graded from 0 to 3 (absent (0), weakly diffuse (1), moderate granular (2), strong granular or diffuse (3)). In the statistical analysis, the COX-2 grades were handled in two subgroups, i.e. negative (scores 0– 1) and positive (scores 2–3). The HuR and COX-2 immunoreactivity was graded by two independent interpreters (HR and AR) who had no preliminary knowledge of the clinical data. The case-by-case final consensus grade was discussed and determined in a common session.

HuR protein plays an important role in carcinogenesis [8] and its cytoplasmic expression has been connected with the poor prognosis of many carcinomas such as ovarian [9, 10], gastric [11] and breast carcinomas [12]. One mechanism by which HuR may facilitate its carcinogenic effect is by the induction of cyclooxygenase-2 (COX-2) expression [9–11, 13]. Cyclooxygenase-2 has been linked to carcinogenesis via the modulation of tumour cell growth and apoptosis, angiogenesis, invasion and metastasis [14], and its expression has been shown to be associated with reduced survival in RCC patients [15]. Since the prognostic significance of HuR in RCC has not been evaluated, we investigated the association between HuR and COX-2 expression and clinicopathological variables, including disease-specific survival in RCC specimens using immunohistochemistry.

Tumour samples The tumour samples were fixed in 10% buffered formalin and embedded in paraffin. The histological diagnosis was confirmed by reviewing haematoxylin and eosin (H & E)stained original sections, simultaneously by two pathologists. The tumours were re-classified and graded according to the WHO classification [18]. The most representative block was selected and cut into 3-μm thick sections in multi-tissue blocks which were mounted onto pre-coated

Cell culture and RNA interference Human RCC cells (769-P, ATCC CRL-1933) were maintained in RPMI 1640 supplemented with 10% fetal calf serum (FCS) (Sigma-Aldrich, St. Louis, MO, USA), 2 mmol/l L-glutamine and antibiotics (Sigma-Aldrich) at 37°C and 5% CO2. The small interfering RNA (siRNA) duplexes were synthesised by Dharmacon, Inc. (Lafayette, CO, USA) and the sequences for HuR were sense 5′AACAUGACCCAGGAUGAGUUA-dTdT-3′ and antisense 5′-UAACUCAUCCUGGGUCAUGUU-dTdT-3′. ON-TARGETplus Non-Targeting Pool (D-001810-10, Dharmacon) was used as a control for the siRNA experiments. The day before transfection, 769-P cells were

Tumor Biol. (2011) 32:481–487

trypsinized and diluted at 1:20 with optiMEM 1 medium (Life Technologies, Inc., Paisley, United Kingdom) supplemented with 10% FCS without antibiotics and transferred to 24-well plates at 1 ml per well with the final split ratio being 1:4. Transient transfection of siRNAs was carried out using lipofectamine reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), following the instructions of the manufacturer. The final small interfering RNA concentrations were 80 nmol/l. Proteins from the cells were isolated for 78 h after transfection using RIPA buffer (25 mM Tris– HCl pH 7.6, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS) as previously described [19]. After boiling the samples in reducing SDS sample buffer for 5 min, 20 μg of protein was loaded per lane and the samples were separated by 10% SDS-PAGE.

483

HuR immunostaining. For further analysis, the weakly and strongly positive cases were combined and regarded as positive (n=37, 25%). Nuclear HuR immunopositivity was detected in 134 (88%) tumour samples. Cytoplasmic COX2 immunostaining was negative in 103 (70%) of the tumours, weakly positive in 22 (15%), moderately positive in ten (7%) and strongly positive in 12 (8%) of the tumours; COX-2 immunostaining was not detected in the nucleus. For further analysis, the subgroups of negative and weakly positive were combined and regarded as negative (n=125, 85%), and moderately and strongly positive ones as positive (n=22, 15%). Cytoplasmic HuR and COX-2 expression had a positive association, p<0.001. Both HuR and COX-2 expressions were present in 16 (11%) tumours, whereas no HuR or COX-2 expression was detected in 102 (70%) tumours.

Immunoblotting Proteins from the SDS gel were transferred to a PVDF membrane for HuR (sc-56709, Santa Cruz Biotechnology), COX-2 (sc-1745, Santa Cruz Biotechnology) and GAPDH (SM7116P, Acris Antibodies GmbH, Herford, Germany) detection. Binding of the target proteins onto the membranes was revealed with goat anti-mouse or donkey antigoat HRP-conjugated secondary antibodies (Santa Cruz Biotechnology). The protein bands were visualised by chemiluminescence. Statistical analyses The software SPSS for Windows 15 (Chicago, IL) was used for statistical analyses. Associations between factors, including clinicopathological variables and HuR and COX-2 immunostaining patterns, were assessed by the χ2 test or the Fisher’s exact test in the case of low expected frequencies. Survival rates were calculated using the Kaplan–Meier method, and the statistical significance between groups was analysed using the log-rank test. Renal cell carcinoma-specific survival was calculated from the date of diagnosis to death from RCC or the last day of follow-up. Deaths due to intercurrent causes were censored. Multivariate survival analysis was done with the Cox proportional hazards model; the following covariates were entered: stage, Fuhrman grade, HuR immunoreactivity and COX-2 immunoreactivity. All p values were two sided.

Association of cytoplasmic HuR and COX-2 expression with the clinicopathological characteristics The distributions of pT-class, stage, and nuclear grade and their associations with cytoplasmic HuR expression are presented in Table 1. Cytoplasmic HuR expression was associated with advanced stage (p=0.004). Tumours with HuR expression showed a tendency towards a higher Fuhrman grade but without statistical significance (p= 0.06). Furthermore, there was a significant association between histological subtype of RCC and cytoplasmic HuR expression (p=0.01). Most of the clear cell and chromophobic carcinomas (n=100 (78%) and n=4 (80%), respectively) did not express HuR, whereas papillary carcinomas more frequently showed HuR expression (n=5 (56%) positive). Both of the unclassified tumours expressed HuR. Nuclear HuR expression did not have any association with these characteristics (data not shown). As shown in Table 1, cytoplasmic COX-2 expression was associated with high pT-class, advanced stage (p=0.04) and high nuclear grade (p=0.04). Cytoplasmic expression of COX-2 was also associated with histological type (p=0.007). As with HuR, the majority of clear cell carcinomas did not express COX-2 (n=115, (88%)), whereas COX-2 expression was more frequently detected in papillary (n=3 (33%)), chromophobic (n=1 (20%)) and unclassified (n=2 (100%)) carcinomas. Prognostic significance of HuR and COX-2 expression in RCC

Results HuR and COX-2 protein expression in RCC Six (4%) of the tumours were strongly positive, 31 (21%) weakly positive and 108 (75%) negative for cytoplasmic

The RCC-specific survival was longer for patients whose tumours did not express HuR (p=0.012) (Fig. 1a). When only clear cell carcinomas were included in the survival analysis, the lack of HuR expression was an even stronger prognostic factor of RCC-specific survival (p=0.006). For

484 Table 1 Associations between HuR and COX-2 expression and tumour pT-class, stage and grade

Tumor Biol. (2011) 32:481–487 Cytoplasmic HuR expression Negative pT-class pT1 59 (81%) pT2 5 (45%) pT3 41 (75%) pT4 3 (50%) stage I 56 (82%) II 5 (45%) III 38 (79%) IV 9 (50%) Nuclear grade I 4 (80%) II 66 (83%) III 27 (71%) IV 11 (55%)

Positive

p value

Negative

Positive

14 6 14 3

(19%) (55%) (25%) (50%)

0.04

67 9 46 3

(92%) (82%) (81%) (50%)

6 2 11 3

(8%) (18%) (19%) (50%)

0.03

12 6 10 9

(18%) (55%) (21%) (50%)

0.004

63 9 41 12

(93%) (82%) (82%) (67%)

5 2 9 6

(7%) (18%) (18%) (33%)

0.04

1 14 11 9

(20%) (17%) (29%) (45%)

0.06

5 74 29 17

(100%) (93%) (76%) (77%)

0 6 9 5

(0%) (7%) (24%) (23%)

0.04

patients who died within the first 5 years after diagnosis, the mean RCC-specific survival times with and without HuR expression in their tumours were 46 (95% CI, 39–53) and 54 (95% CI, 0–57) months, respectively, with p=0.006. The hazard ratio (HR) of patients with HuR-expressing tumours was 2.18 (95% CI, 1.16–4.09, p=0.015). Nuclear HuR expression was not associated with RCC-specific survival (p=0.2). Lack of COX-2 expression was a marker of better prognosis (p=0.015) (Fig. 1b). During the 5-year followup, the RCC-specific mean survival times for patients with and without COX-2 expression in their tumours were 42 (95% CI, 33–50) and 53 months (95% CI, 50–56), respectively (p=0.007). The HR of patients with COX-2expressing tumours was 2.29 (95% CI, 1.15–4.54, p= 0.018). In the Cox regression analysis for cytoplasmic HuR and COX-2 expression, stage and nuclear grade, the only statistically significant factor in RCC-specific survival was

Fig. 1 Associations of cytoplasmic HuR and COX-2 immunopositivity in RCC with RCCspecific survival. Patients with HuR (a) and COX-2 (b) positive tumours showed reduced survival when compared to patients with tumours negative for these proteins

Cytoplasmic COX-2 expression p value

stage (p<0.001) (Table 2). HRs of patients with stage II, stage III and stage IV disease were 5.12 (95% CI, 1.62– 16.2, p=0.005), 3.80 (1.56–9.24, p=0.003) and 24.1 (9.61– 60.4, p<0.001), respectively. HRs of patients with Fuhrman grade III and Fuhrman grade IV tumours were 1.76 (95% CI, 0.86–3.63, p=0.12) and 3.92 (1.90–8.09, p<0.001), respectively. Fuhrman grade I and grade II tumours were combined for reference class due to low number of grade I tumours. Effect of HuR siRNA on COX-2 expression in RCC cells To test the hypothesis that HuR regulates COX-2 expression in RCC cells, we treated 769-P cells with HuR siRNA molecules. This treatment suppressed HuR expression, as detected by immunoblotting (Fig. 2). Furthermore, we observed a suppression of COX-2 protein expression after the HuR siRNA treatment (Fig. 2). The control siRNA showed no effect on HuR or COX-2 expression. These

Tumor Biol. (2011) 32:481–487 Table 2 Cox multivariate survival analysis in 143 patients with RCC

485 Covariate

Hazard ratio

Stage I Stage II Stage III Stage IV Fuhrman grade I or II Fuhrman grade III Fuhrman grade IV Cytoplasmic HuR expression Cytoplasmic COX-2 expression

1 (ref.) 3.78 3.25 21.2 1 (ref.) 0.78 1.42 1.45 1.08

results indicate that HuR can regulate COX-2 expression in RCC cells.

Discussion In the search for novel prognostic markers for RCC, we evaluated the significance of HuR and COX-2 expression in RCC tumours in our cohort of surgically treated RCC patients. Cytoplasmic HuR expression was associated with reduced RCC-specific survival. A similar association between HuR expression and prognosis has also been detected in several other carcinomas as well [9, 11, 12]. Our clinical data presented here is supported by experimental findings presented previously, which may explain the less favourable outcome of RCC patients with cytoplasmic HuR expression. Knockdown of HuR in an RCC cell line (786-O) led to decreased proliferation [20] and an increase in apoptosis [21], which was also shown in an in vivo tumour model using subcutaneous tumour inoculation in nude mice followed by injections of HuR siRNAs. Treatment of the tumours with HuR siRNAs led to the inhibition of cell proliferation and an induction of cell apoptosis [21]. The same authors documented a greater expression of HuR in the cytoplasmic fraction of pT1 (n=7) tumours compared to the expression in pT3 (n=8) tumours, as measured by immunoblotting, which the authors suggested as indicating a role of HuR in tumour promotion [21]. In this study, we detected a significant association

Fig. 2 Effect of HuR siRNA on the expression of COX-2 in RCC cells. HuR-targeted siRNA inhibited HuR and COX-2 protein expression in 769-P RCC cells, as detected by immunoblotting (lane 1). Control siRNA did not have an effect on HuR or COX-2 expression (lane 2). Untreated 769-P cells (lane 3). Glyceraldehyde3-phosphate dehydrogenase (GAPDH) was used as a loading control

95.0% CI

p value

1.08–13.3 1.28–8.24 7.44–60.6

0.038 0.013 <0.001

0.35–1.72 0.60–3.37 0.61–3.44 0.41–2.86

0.54 0.42 0.40 0.88

between HuR and COX-2 expression. The HuR protein binds to COX-2 mRNA in the nucleus, after which the complex is transported to the cytoplasm [9]. Three binding sites have been shown for HuR in the 3′-untranslated region of human COX-2, and the binding of HuR to COX-2 seems to be important for post-transcriptional COX-2 mRNA stabilisation [22]. In ovarian cancer and gastric cancer, one plausible explanation for the prognostic significance of HuR may be related to the ability of HuR to induce COX-2 expression [9, 11], which may also be present in RCC. Indeed, we were able to detect a similar connection between HuR and COX-2 expression in the RCC cell line, since HuR knockdown by siRNA led to the reduced expression of COX-2. The expression of COX-2 was shown to be associated with an advanced stage of RCC and other clinicopathological features such as tumour size and grade, but it is not an independent prognostic factor in RCC [15, 23–25]. Our results are in line with these previous results, as here COX-2 expression was associated with high nuclear grade and high clinical stage. Both nuclear and cytoplasmic expressions of HuR have previously been reported in RCC [21, 26], which is in line with our results. In colorectal carcinomas, the expression of HuR was mainly nuclear, but some expression in the nonnuclear compartment was also detected [27]. The antibody used by Yoo and colleagues [27] was not the same as mostly used anti-human mouse monoclonal clone 19F12 antibody, which was also used in this study. In ovarian carcinomas, a dominant nuclear staining of HuR has been reported with conditions for immunohistochemistry optimised for the nuclear staining of HuR [28]. In the prostate, nuclear localization of HuR was detected in benign tissue with a transition to cytoplasmic expression in prostate cancer [29]. In this study, we could not detect any prognostic significance or associations with clinical characteristics for nuclear HuR expression. There are several potential mechanisms which explain the less favourable outcome of RCC patients with cytoplasmic HuR expression, in addition to COX-2 regulation. The parathyroid hormone-related protein (PTHrP) is a

486

prognostic factor in renal cell carcinoma [30], and blocking of PTHrP induces cell death in RCC in vitro and in animal models [31]. It was recently shown that one of the proteins binding to PTHrP mRNA is HuR, and HuR knockdown decreased the expression of PTHrP [20]. VEGF is a key mediator of angiogenesis for both physiological and pathological conditions [32]. It has been demonstrated that HuR is important for VEGF mRNA stabilisation [33]. The stability of VEGF mRNA is regulated by HuR in hypoxic cells [34]. Further, the over-expression of HuR results in the elevated expression of VEGF protein [13]. Furthermore, HuR can bind to the 3′ untranslated region of the p53 mRNA in a wild-type VHL protein (pVHL)-dependent fashion, and HuR knockdown decreased the translation and abundance of p53 in pVHL-expressing cells [35]. Also, pVHL reduces the stability of type I insulin-like growth factor receptor (IGF1R) mRNA via sequestration of HuR protein [36], and IGF1R, in turn, is one of the key mediators of tumour cell survival. Expression of IGF1R has been connected to a worse prognosis in RCC [37]. Finally, HuR knockdown decreased the activation of the oncogenic PI3K/Akt and MAPK pathways in RCC [21]. We studied HuR and COX-2 expression by immunohistochemistry only, which is a limitation of the present study. However, immunohistochemistry detects the expression of the functional gene products, and measuring RNA expression levels are not always congruent with those of the protein. To this end, since our results show significant associations of HuR and COX-2 protein expression with prognosis of RCC, our data are important and suggest a novel link between the mRNA stabilising HuR and carcinogenesis of the kidney. We conclude that the cytoplasmic expression of HuR is associated with a reduced RCC-specific survival. One potential explanation for the greater rate of fatal disease among patients with an increased HuR expression is that HuR mediates COX-2 expression, as the expression of HuR correlated with COX-2 expression. In addition, cell line experiments proved that the reduction in HuR expression leads to the inhibition of COX-2 expression. Thus, HuR seems to be one of the key players in RCC carcinogenesis and further studies are needed to elucidate its prognostic significance in other RCC patient cohorts. Since a prognostic significance of HuR immunostaining was demonstrated by a simple negative vs. positive cytoplasmic immunostaining stratification, this could be a potential prognostic marker for clinical use. Acknowledgements We thank Ms. Mirja Vahera, Ms. Erja Tomperi and Ms. Mirja Mäkeläinen for their skilful technical assistance. HR received grants from the Cancer Association of Northern Finland and from the Finnish Urological Association for this study. MHV was supported by the Finnish Medical Fund.

Tumor Biol. (2011) 32:481–487

References 1. Pantuck AJ, Zisman A, Belldegrun AS. The changing natural history of renal cell carcinoma. J Urol. 2001;166:1611–23. 2. Breda A, Konijeti R, Lam JS. Patterns of recurrence and surveillance strategies for renal cell carcinoma following surgical resection. Expert Rev Anticancer Ther. 2007;7:847–62. 3. Phuoc NB, Ehara H, Gotoh T, Nakano M, Kamei S, Deguchi T, et al. Prognostic value of the co-expression of carbonic anhydrase IX and vascular endothelial growth factor in patients with clear cell renal cell carcinoma. Oncol Rep. 2008;20:525–30. 4. McGuire BB, Fitzpatrick JM. Biomarkers in renal cell carcinoma. Curr Opin Urol. 2009;19:441–6. 5. Ma WJ, Cheng S, Campbell C, Wright A, Furneaux H. Cloning and characterization of HuR, a ubiquitously expressed elav-like protein. J Biol Chem. 1996;271:8144–51. 6. Fan XC, Steitz JA. Overexpression of HuR, a nuclear-cytoplasmic shuttling protein, increases the in vivo stability of ARE-containing mRNAs. EMBO J. 1998;17:3448–60. 7. Brennan CM, Steitz JA. HuR and mRNA stability. Cell Mol Life Sci. 2001;58:266–77. 8. Abdelmohsen K, Gorospe M. Posttranscriptional regulation of cancer traits by HuR. Wiley Interdisciplinary Reviews—RNA; 2010. doi:10.1002/wrna.4. 9. Erkinheimo TL, Lassus H, Sivula A, Sengupta S, Furneaux H, Hla T, et al. Cytoplasmic HuR expression correlates with poor outcome and with cyclooxygenase 2 expression in serous ovarian carcinoma. Cancer Res. 2003;63:7591–4. 10. Denkert C, Weichert W, Pest S, Koch I, Licht D, Kobel M, et al. Overexpression of the embryonic-lethal abnormal vision-like protein HuR in ovarian carcinoma is a prognostic factor and is associated with increased cyclooxygenase 2 expression. Cancer Res. 2004;64:189–95. 11. Mrena J, Wiksten JP, Thiel A, Kokkola A, Pohjola L, Lundin J, et al. Cyclooxygenase-2 is an independent prognostic factor in gastric cancer and its expression is regulated by the messenger RNA stability factor HuR. Clin Cancer Res. 2005;11:7362–8. 12. Heinonen M, Bono P, Narko K, Chang SH, Lundin J, Joensuu H, et al. Cytoplasmic HuR expression is a prognostic factor in invasive ductal breast carcinoma. Cancer Res. 2005;65:2157– 61. 13. Dixon DA, Tolley ND, King PH, Nabors LB, McIntyre TM, Zimmerman GA, et al. Altered expression of the mRNA stability factor HuR promotes cyclooxygenase-2 expression in colon cancer cells. J Clin Invest. 2001;108:1657–65. 14. Menter DG, Schilsky RL, DuBois RN. Cyclooxygenase-2 and cancer treatment: understanding the risk should be worth the reward. Clin Cancer Res. 2010;16:1384–90. 15. Miyata Y, Koga S, Kanda S, Nishikido M, Hayashi T, Kanetake H. Expression of cyclooxygenase-2 in renal cell carcinoma: correlation with tumor cell proliferation, apoptosis, angiogenesis, expression of matrix metalloproteinase-2, and survival. Clin Cancer Res. 2003;9:1741–9. 16. Ronkainen H, Kauppila S, Hirvikoski P, Vaarala MH. Evaluation of myosin VI, E-cadherin and beta-catenin immunostaining in renal cell carcinoma. J Exp Clin Cancer Res. 2010;29:2. 17. UICC. TNM classification of malignant tumours. 6th ed. New York: Wiley; 2002. 18. IARC. Tumours of the urinary system and male genital organs. Lyon: IARC; 2004. 19. Sefton BM. Labeling cultured cells with 32Pi and preparing cell lysates for immunoprecipitation. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K, editors. Current Protocols in Molecular Biology. London: Wiley; 2005. Unit 18.2.

Tumor Biol. (2011) 32:481–487 20. Danilin S, Sourbier C, Thomas L, Rothhut S, Lindner V, Helwig JJ, et al. Von hippel-lindau tumor suppressor gene-dependent mRNA stabilization of the survival factor parathyroid hormonerelated protein in human renal cell carcinoma by the RNA-binding protein HuR. Carcinogenesis. 2009;30:387–96. 21. Danilin S, Sourbier C, Thomas L, Lindner V, Rothhut S, Dormoy V, et al. Role of the RNA-binding protein HuR in human renal cell carcinoma. Carcinogenesis. 2010;31:1018–26. 22. Sengupta S, Jang BC, Wu MT, Paik JH, Furneaux H, Hla T. The RNA-binding protein HuR regulates the expression of cyclooxygenase-2. J Biol Chem. 2003;278:25227–33. 23. Hashimoto Y, Kondo Y, Kimura G, Matsuzawa I, Sato S, Ishizaki M, et al. Cyclooxygenase-2 expression and relationship to tumour progression in human renal cell carcinoma. Histopathology. 2004;44:353–9. 24. Tuna B, Yorukoglu K, Gurel D, Mungan U, Kirkali Z. Significance of COX-2 expression in human renal cell carcinoma. Urology. 2004;64:1116–20. 25. Sozen S, Gurocak S, Erdem O, Acar C, Kordan Y, Akyol G, et al. Cyclooxygenase-2 expression: does it have a probable role in tumorigenesis mechanisms of renal cell carcinoma? Int Urol Nephrol. 2008;40:295–301. 26. Lopez de Silanes I, Lal A, Gorospe M. HuR: post-transcriptional paths to malignancy. RNA Biol. 2005;2:11–3. 27. Yoo PS, Sullivan CA, Kiang S, Gao W, Uchio EM, Chung GG, et al. Tissue microarray analysis of 560 patients with colorectal adenocarcinoma: high expression of HuR predicts poor survival. Ann Surg Oncol. 2009;16:200–7. 28. Yi X, Zhou Y, Zheng W, Chambers SK. HuR expression in the nucleus correlates with high histological grade and poor diseasefree survival in ovarian cancer. Aust N Z J Obstet Gynaecol. 2009;49:93–8. 29. Niesporek S, Kristiansen G, Thoma A, Weichert W, Noske A, Buckendahl AC, et al. Expression of the ELAV-like protein HuR

487

30.

31.

32.

33.

34.

35.

36.

37.

in human prostate carcinoma is an indicator of disease relapse and linked to COX-2 expression. Int J Oncol. 2008;32:341–7. Iwamura M, Wu W, Muramoto M, Ohori M, Egawa S, Uchida T, et al. Parathyroid hormone-related protein is an independent prognostic factor for renal cell carcinoma. Cancer. 1999;86:1028– 34. Massfelder T, Lang H, Schordan E, Lindner V, Rothhut S, Welsch S, et al. Parathyroid hormone-related protein is an essential growth factor for human clear cell renal carcinoma and a target for the von hippel-lindau tumor suppressor gene. Cancer Res. 2004;64:180–8. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science. 1989;246:1306–9. Datta K, Mondal S, Sinha S, Li J, Wang E, Knebelmann B, et al. Role of elongin-binding domain of von hippel lindau gene product on HuR-mediated VPF/VEGF mRNA stability in renal cell carcinoma. Oncogene. 2005;24:7850–8. Levy NS, Chung S, Furneaux H, Levy AP. Hypoxic stabilization of vascular endothelial growth factor mRNA by the RNA-binding protein HuR. J Biol Chem. 1998;273:6417–23. Galban S, Martindale JL, Mazan-Mamczarz K, Lopez de Silanes I, Fan J, Wang W, et al. Influence of the RNA-binding protein HuR in pVHL-regulated p53 expression in renal carcinoma cells. Mol Cell Biol. 2003;23:7083–95. Yuen JS, Cockman ME, Sullivan M, Protheroe A, Turner GD, Roberts IS, et al. The VHL tumor suppressor inhibits expression of the IGF1R and its loss induces IGF1R upregulation in human clear cell renal carcinoma. Oncogene. 2007;26:6499–508. Parker AS, Cheville JC, Janney CA, Cerhan JR. High expression levels of insulin-like growth factor-I receptor predict poor survival among women with clear-cell renal cell carcinomas. Hum Pathol. 2002;33:801–5.