J Cancer Res Clin Oncol DOI 10.1007/s00432-015-1942-1
ORIGINAL ARTICLE – CANCER RESEARCH
Functional analyses and prognostic significance of SFRP1 expression in bladder cancer Anja Rogler · Emil Kendziorra · Johannes Giedl · Christine Stoehr · Helge Taubert · Peter J. Goebell · Bernd Wullich · Michael Stöckle · Jan Lehmann · Sabrina Petsch · Arndt Hartmann · Robert Stoehr
Received: 14 January 2015 / Accepted: 16 February 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose We previously showed that the Wnt-signaling antagonist SFRP1 (secreted frizzled-related protein 1) is a promising marker in bladder cancer. The aim of this study was to validate the prognostic role and analyze the functional significance of SFRP1. Methods Four bladder cancer cell lines (RT112, RT4, J82 and BFTC905) and one urothelial cell line (UROtsa) were used for functional characterization of SFRP1 expression. Effects on viability, proliferation and wound healing were investigated, and canonical Wnt-pathway activity as well as
A. Rogler · J. Giedl · C. Stoehr · A. Hartmann · R. Stoehr (*) Institute of Pathology, University Hospital Erlangen, FriedrichAlexander-University Erlangen-Nuremberg, Krankenhausstr. 8‑10, 91054 Erlangen, Germany e-mail: robert.stoehr@uk‑erlangen.de E. Kendziorra Clinical Research Unit 179, Department of General, Visceral and Pediatric Surgery, University Hospital Göttingen, Robert‑Koch‑Str. 40, 37075 Göttingen, Germany H. Taubert · P. J. Goebell · B. Wullich Department of Urology, University Hospital Erlangen, FriedrichAlexander-University Erlangen-Nuremberg, Krankenhausstr. 12, 91054 Erlangen, Germany M. Stöckle Clinic for Urology and Children’s Urology, University Hospital Saarland, Kirrberger Straße, 66421 Homburg, Saar, Germany
Wnt-signaling target gene expression was analyzed. Additionally, tissue micro-arrays from two different bladder tumor cohorts were evaluated for SFRP1 expression, and associations with survival and histopathological parameters were analyzed. Results The cell lines RT112, RT4, J82 and UROtsa showed SFRP1 expression. In BFTC905, SFRP1 expression was inhibited by promoter hypermethylation. Wntpathway activity was absent in all cell lines and independent from SFRP1 expression. RT112 and BFTC905 were used for further functional characterization. SFRP1 overexpression resulted in decreased viability and migration in BFTC905 cells. Knockdown of SFRP1 expression in RT112 cells resulted only in marginal effects. In bladder tumors, SFRP1 expression was associated with lower tumor grade, but not with progression in patients with papillary bladder cancer. SFRP1 expressing papillary bladder cancer tumors also demonstrated a tendency to longer overall survival. Conclusions SFRP1 is reducing malignant potential of BFTC905 cells, but not by regulation of canonical Wntsignaling pathway. Other pathways, like non-canonical Wnt or the MAPK pathway, could be activated via SFRP1expression loss. In bladder tumors, SFRP1 has the potential to predict outcome for a subset of papillary bladder tumors. Keywords Bladder cancer · Canonical Wnt signaling · SFRP1 · TOP flash/FOP flash assay
J. Lehmann Urology Practice Prüner Gang, Prüner Gang 15, 24105 Kiel, Germany
Introduction
S. Petsch Tumor Zentrum, Friedrich-Alexander-University ErlangenNuremberg, Carl‑Thiersch‑Str. 7, 91052 Erlangen, Germany
Bladder cancer was the sixth most common cancer in Europe in 2008 with urothelial cell carcinoma (UCC) representing the predominant form in 90 % of all cases (Ferlay et al.
13
2010). With regard to histopathology, clinical behavior and molecular profile, two major groups of UCC can be distinguished (Wu 2005). Noninvasive papillary low-grade tumors are thought to develop from urothelial hyperplasia and are characterized by chromosome 9p/q-deletions as well as by activating mutations in FGFR3, PIK3CA and HRAS. They tend to recur frequently, but have a relative low progression rate (Knowles 2006, 2008). Invasive bladder cancers are thought to arise via flat dysplasia and carcinoma in situ and are characterized by initiating P53 alterations, followed by extensive genomic instability with multiple chromosomal aberrations resulting in loss of tumor suppressor genes like PTEN and RB (Knowles 2006, 2008). Thus, invasive bladder tumors show a high tendency to progress and to metastasize, require aggressive treatments and are still associated with poor 5-year survival rates (Feifer et al. 2011). To provide the best appropriate treatment for every patient and to predict individual progression risk and outcome, new reliable molecular markers are needed. In previous experiments, we could find an association of chromosome 8p deletions with papillary, muscle-invasive growth pattern in 99 bladder tumors. Further analysis identified the candidate gene secreted frizzled-related protein 1 (SFRP1), which was deleted in 38 % of all bladder cancer cases and which was to some extent a consequence of promoter hypermethylation. This expression loss was associated with inferior outcome in patients with papillary, muscle-invasive bladder cancer (Stoehr et al. 2004). SFRP1 expression loss was reported in many other malignancies (e.g., in colorectal, breast or kidney cancer) and correlated with promoter hypermethylation (Dahl et al. 2007; Tanaka et al. 2008; Veeck et al. 2006). Thus, we hypothesized that SFRP1 might act as a tumor suppressor gene in bladder cancer and that it has the potential to predict disease progression in papillary bladder cancer. SFRP1 (NCBI Gene ID 6422) is located at chromosome 8p12-11.1 and is transcribed into a secreted 35 kDa glycoprotein. SFRP1 can act as Wnt-signaling inhibitor, via binding of extracellular Wnt-ligands or direct binding of transmembranous receptor FRIZZLED (FZD) (Finch et al. 1997). The aim of this project was to validate the role of SFRP1 as potential progression marker in bladder cancer using cell culture experiments on the one hand and tumor samples on the other. We sought to clarify functional effects of SFRP1 expression in cell lines using viability, proliferation and wound-scratch assays. To investigate consequences on Wntsignaling activity, we performed TOP flash/FOP flash assay and Wnt-target gene expression analysis of AXIN2, BMP4, CD44, CMYC, CYCLIN D1 and SURVIVIN. Furthermore, SFRP1 expression was investigated immunohistochemically in two different bladder cancer cohorts with 86 and 243 patients, respectively, and correlated with survival data.
13
J Cancer Res Clin Oncol
Materials and methods Patient cohorts and tumor specimen Tissue micro-arrays (TMAs) of two different bladder cancer patient cohorts were used for immunohistochemical analysis of SFRP1: group 1 consisted of 86 patients with non-muscle-invasive (pTa or pT1) papillary tumors and group 2 of 243 patients with advanced bladder tumors (≥pT3 and/or pN1), who all underwent radical cystectomy and received adjuvant chemotherapy. TMAs of the advanced tumor group were constructed at the Institute of Pathology Erlangen after collecting the paraffin blocks from a previous prospective adjuvant chemotherapy study of advanced bladder cancers (Lehmann et al. 2005) originally consisting of 327 patients. Due to tissue availability, only a subgroup of 243 patients (74 %) of the initial cohort could be analyzed. From all blocks, HE sections were made and all histopathological parameters were reviewed by one pathologist experienced in uro-pathology. For this study, IRB approval was obtained from the German Association of Urological Oncology (AUO) as well as informed written consent was obtained from all patients of participating local centers and clinics. All relevant patient characteristics and clinico- and histopathological parameters were summarized previously (Lehmann et al. 2005). Papillary bladder tumors were newly collected for this study from the tumor bank of the Comprehensive Cancer Center Erlangen-EMN located at the Institute of Pathology in Erlangen. Formalin-fixed and paraffin-embedded tumor tissues and corresponding hematoxylin–eosin stained sections were selected, and tumor areas were marked and reevaluated according to histopathological stage and grade by two experienced surgical pathologists (AH, JG). The patient cohort was aged between 27 and 97 years (mean age: 70.3 ± 11.4 years) and consisted of 64 men and 22 women. Clinical follow-up data for the papillary tumor group were obtained in collaboration with the clinical cancer registry (Tumorzentrum, TUZ) of Erlangen. Informed written consent was obtained from all patients of the papillary tumor group, and IRB approval was obtained from local ethics committee of the University Clinic Erlangen for retrospective use of patient material in the context of the Comprehensive Cancer Center tissue bank. All relevant patient characteristics, histopathological data and follow-up are shown in Table 1. Cell lines and transfections For functional analysis of SFRP1-expression, the bladder cancer cell lines RT112, RT4, J82 and BFTC905 as well as the normal urothelial cell line UROtsa were screened
J Cancer Res Clin Oncol Table 1 Patient characteristics of bladder cancer cohorts
Patients Age
Gender
Papillary bladder tumor cohort
Advanced bladder tumor cohort
n = 86 Mean: 70.3 years Median: 72 years (±11.4 years) Range: 29–97 years
n = 243 Mean: 62.1 years Median: 63 years (±8.3 years) Range: 38–81 years
Female: n = 22
Female: n = 56
Male: n = 64
Male: n = 182 n.a.: n = 5
Stage
PUNLMP n = 1
pT1, pN+ n = 5
pTa n = 47
pT2, pN+ n = 28
pT1 n = 32
pT3 n = 144
pT2 n = 4
pT4 n = 40
pT3 n = 1
n.a. n = 26
pT4 n = 1 Grade
Follow-up OS
Follow-up DSS
lg n = 40
G2, hg n = 27
hg n = 45
G3, hg n = 209
PUNLMP n = 1
n.a. n = 7
Alive n = 66
Alive n = 130
Dead n = 15 n.a. n = 5
Dead n = 80 n.a. = 34
Alive n = 70
Alive n = 144
Dead n = 8
Dead n = 66
n.a. n = 8
n.a. = 43
using conventional PCR, qRT-PCR, Western blot analysis and immunofluorescence, respectively (Hatina et al. 2008; Masters et al. 1986; Rieger et al. 1995; Rossi et al. 2001; Tzeng et al. 1996). The malignant melanoma cell line SKMel-28 (Carey et al. 1976) was used as positive control for SFRP1-protein detection, and colorectal carcinoma cell line SW480 (Leibovitz et al. 1976) was used as positive control in TOP flash/FOP flash assay. Cells were cultured in RPMI medium supplemented with 10 % fetal calf serum (FCS), 1 % sodium-pyruvate and 1 % l-glutamine at 37 °C and 5 % CO2. Transfections were carried out in 6-well plates seeding 100,000 and 300,000 cells per well for RT112 and BFTC905, respectively. After 24 h of cell adhesion, SFRP1 was stable overexpressed in BFTC905 using the expression plasmid for human SFRP1 in pCMV6-Neo vector (Origene Technologies, Rockville, USA) and FuGene HD (Roche, Penzberg, Germany) with a ratio of 2:6 (DNA:FuGene) according to the manufacturer’s instructions. For clonal selection, 250 µg/µl G-418 (Neomycin, Roche) was added to the culture medium. After 36 h of cell adhesion, SFRP1 was stable knocked down in RT112 using the anti-SFRP1 shRNA overexpression plasmid in pGFP-V-RS shRNA vector (Origene
Technologies) and TurboFectin 8.0 (Roche) with a ratio of 1:3 (DNA/TurboFectin) according to the manufacturer’s instructions. For clonal selection, 3 µg/ml puromycin (InvivoGen, San Diego, USA) was added to the culture medium. To reverse hypermethylation and acetylation, 200,000 cells were cultured in 3 ml RPMI medium, and 2 µM 5-Aza-2′deoxycytidine (Sigma-Aldrich, Taufkirchen, Germany), 2 µM suberoylanilide hydroxamic acid (SAHA, Biozol, Eching, Germany) and 1.65 µM trichostatin A/vorinostat (Sigma-Aldrich) were added. Isolation of DNA and RNA and cDNA synthesis To investigate SFRP1 promoter methylation in papillary urothelial carcinomas and cell lines, tumor specimens were manually micro-dissected, cells were pelleted and DNA was isolated using the High Pure PCR Template Preparation Kit (Roche) according to the manufacturer’s protocol. To analyze SFRP1 and Wnt-target gene expression with nested and qRT-PCR, respectively, RNA was isolated using RNeasy® Mini Kit (Qiagen, Hilden, Germany) and cDNA was converted using the RevertAid™ H Minus First Strand cDNA Synthesis Kit (Fermentas Life Sciences, St. Leon-Rot, Germany) according to the manufacturer’s instructions. For cDNA synthesis, 1 µg total RNA was used. DNA- and RNA-quality was controlled using the Multiplate Reader Synergy 2 (BioTek, Bad Friedrichshall, Germany). Nested PCR Genetic region of SFRP1 was amplified by nested PCR using two pairs of primers (outer sense: 5′-CAC AACGTGGGCTACAAGAA-3′, outer antisense: 5′-GAAG TGGTGGCTGAGGTTGT-3′, inner sense: 5′-CTACTGG CCCGAGATGCTTA-3′, inner antisense: 5′-GCTGGCAC AGAGATGTTCAA-3′) obtained from Metabion (Martinsried, Germany) in a total volume of 25 µl containing approx. 100 ng DNA, 0.2 mM dNTP (Promega, Mannheim, Germany), 0.18 µM primers, 5 µl 5× buffer (Promega) and 0.0025 U/µl GoTaq (Promega). For the nested PCR, two PCRs were carried out using the outer primer pair and additional 5 % DMSO in the first reaction and using the inner primer pair and 2 µl PCR product from the first reaction as template in the second PCR. The thermal cycling conditions for both reactions were as follows: initial denaturation for 3 min at 95 °C, 35 cycles of denaturation at 95 °C for 1 min, annealing at 62 °C (first PCR) or 60 °C (second PCR) for 1 min, elongation at 72 °C for 1 min and final primer extension at 72 °C for 10 min. GAPDH was used as endogenous control, and for its sufficient amplification only the first PCR was necessary using the following primers (Metabion):
13
J Cancer Res Clin Oncol
sense: 5′-TGGTCACCAGGGCTGCTT-3′ and antisense: 5′-GTCTTCTGGGTGGCAGTGAT-3′. Finally, 15 µl of SFRP1 PCR product and 5 µl of GAPDH PCR product were loaded onto 2.5 % agarose gel. The gel was stained for 20 min with ethidiumbromide and visualized under UV-illumination.
at 60 °C. Data analysis was performed with 7500 Software v2.0.5 (Applied Biosystems), and gene expression ratios were calculated with ΔΔCT-method (Schmittgen and Livak 2008).
Bisulfite treatment and methylation‑specific PCR
Immunohistochemistry was performed on formalin-fixed, paraffin-embedded (FFPE) 4 µm TMA sections of tumor tissue specimens transferred to glass slides. Before using TMAs for expression analysis, whole-bladder tumor sections were stained to avoid a bias of the results by a heterogeneous SFRP1 expression in bladder tumors. SFRP1positive tumors showed a diffuse but not a heterogeneous expression throughout the whole tumor (data not shown). TMAs were stained with rabbit anti-SFRP1 antibody (custom-made, Pineda, 1:600) as described below. Briefly, after deparaffinization, endogenous peroxidase was blocked by incubating in 3 % H2O2 for 10 min followed by a 1-min incubation step with citric acid buffer (pH 6.0) at 120 °C to break crosslinks between proteins. Slides were incubated overnight at room temperature with SFRP1 primary antibody. This was followed by incubation with secondary goat anti-rabbit antibody (1:200, Vector Laboratories, Burlingame, USA) for 30 min at room temperature. Then, slides were incubated for 20 min with ABC solution (antibody–biotin complex VECTASTAIN® Elite ABC kit, Vector Laboratories), followed by 10-min incubation with TSAsolution (TSA™ indirect, Perkin Elmer, Waltham, Massachusetts) and 20-min incubation with ABC according to the manufacturer’s protocols. AEC solution (AEC Peroxidase Substrate Kit, Vector Laboratories) was added until staining intensity was sufficient (approx. 10 min). Slides were counterstained for 2 min with hemalaun (Carl Roth, Karlsruhe, Germany) and mounted with Aquatex (Merck, Darmstadt, Germany). Stainings were examined and evaluated by two experienced uropathologist (JG, AH), and immunoreactivity (IRS = immune reactive score) was scored as follows: Intensity (0 = negative, 1 = weak, 2 = moderate, 3 = strong) and number of tumor cells (in percent) were determined. Number of stained cells was correlated with numbers from 0 to 4. No staining of cells was evaluated as 0, 1–25 % as 1, 26–50 % as 2, 51–75 % as 3 and 76–100 % as 4. Numbers were multiplied with staining intensity, and immunoreactive values between 0 and 12 were created. For SFRP1 staining, two immunoreactive groups were made: group 1 = IRS 0 and 1, group 2 = IRS 2–12.
According to the manufacturer’s protocol, 1 µg DNA was used for bisulfite treatment using the EpiTect® Bisulfite Kit (Qiagen); 2 µl bisulfite-treated DNA was used in two methylation-specific PCRs (MSP): one with primers for methylated SFRP1 promoter (“M-PCR”: sense 5′-TGT AGTTTTCGGAGTTA GTGTCGCGC-3, antisense 5′-CCT ACGATCGAAAACGACGCGAACG-3′, Metabion) and one with primers that detected unmethylated SFRP1-promoter (“UM-PCR”: sense: 5′-GTTTTGTAGTTTTTGGAG TTAGTGTTGTGT-3′, antisense: 5′-CTCAACCTACAA TC AAAAACAACACAAACA-3′, Metabion). PCRs were carried out using 0.16 mM dNTPs, 1.2 µM Primermix, 5 % DMSO, 0.0025 U/µl GoTaq and 5 µl 5× buffer in a total volume of 25 µl. In the “unmethylated PCR,” additionally 1 mM MgCl2 was added. The thermal cycling conditions were as follows: initial denaturation for 5 min at 95 °C, 37 cycles of denaturation at 94 °C for 30 s, annealing at 66.5 °C (“M-PCR”) or 63.8 °C (UM-PCR) for 30 s, elongation at 72 °C for 30 s (“M-PCR”) or 45 s (“UM-PCR”) and final primer extension at 72 °C for 10 min. In the “UMPCR,” GoTaq was added after initial denaturation step; 5 µl of each PCR product was loaded onto 2.5 % agarose gel, stained with ethidiumbromide and visualized under UV-illumination. qRT‑PCR To verify SFRP1 wild-type mRNA expression in cell lines and to control knockdown and overexpression of SFRP1 in RT112 and BFTC905, SYBR Green-based quantitative realtime PCR (qRT-PCR) was performed in 7500 Fast (96-well) Real-time PCR system (Applied Biosystems, Darmstadt, Germany) using 25 ng cDNA template, 100nM SFRP1Primermix (sense: 5′-CTACTGGCCCGAGATGCTTA-3′, antisense: 5′-GCTGGCACAGAGATGTTCAA-3′), 250nM GAPDH-Primermix (sense: 5′-TGGTCACCAGGGCTG CTT-3′, antisense: 5′-AGCTTCCCGTTCTCAGCC-3′) and 6.25 µl SYBR Green Mix (2x) in a total volume of 12.5 µl. From SK-Mel-28 cDNA (SFRP1 positive control), 6.25 ng was used. Thermal cycling conditions were as follows: 2 min 50 °C, 10 min 95 °C, 40 cycles of denaturation at 95 °C for 15 s and combined annealing and elongation at 60 °C for 30 s. PCR was completed with melt curve generation: 15 s at 95 °C, 1 min at 60 °C, 30 s at 95 °C and 15 s
13
Immunohistochemistry
Western blotting To analyze SFRP1 and Wnt-target-protein expression in cell lines, we performed Western blot. Therefore, we isolated whole protein fraction from adherent cells incubating
J Cancer Res Clin Oncol
protein lysis buffer (5 M NaCl, 1 M Tris/HCl pH7.2, 1× Triton X, 0.05 % SDS, 0.5 M EDTA pH 8.0, 2 M DTT) on ice for 15 min in the culture flask and cell buffer suspension for 45 min at 4–8 °C. This was followed by ultrasound sonification for 20 s (1 s on/1 s off, 30 % intensity) with HTU SONI 130 (Heinemann, Gmünd, Germany) and centrifugation for 30 min at 4 °C at 13,200g. Protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Bonn, Germany) according to the manufacturer’s protocol. SDS–PAGE was performed with 10 % PAA-gels and 30 µg total protein. 1× Loading buffer (5×: 0.313 M Tris HCl pH 6.8, 10 % SDS (w/v), 0.05 % (w/v) bromphenol blue and 50 % (v/v) glycerol) and 1× DTT (from 20× = 2 M DTT) were added, and lysates were denatured for 5 min at 95 °C. Polyacrylamide gels were run for 15 min at 100 V and then for 60 min at 150 V. Proteins were blotted on nitrocellulose membrane using wet blotting method with Mini Protean® Tetra System (BioRad Laboratories, Munich, Germany) according to the manufacturer’s protocol (25 V, overnight). Membrane was washed with TBS/Tween and blocked for 40 min with BSA (Carl Roth, Karlsruhe, Germany, for AXIN2, BMP4, CD44, CMYC, SFRP1 and SDHA), nonfat dry milk (Carl Roth, for SURVIVIN and ß-AKTIN) or Immunoblot Blocking Reagent (Millipore, Billerica, USA for CYCLIN D1), followed by 1-h incubation with primary antibodies against AXIN2 (rabbit, BioMol/Thermo Fisher Scientific, Bonn, Germany, MA5-15015, 1:133, RT), BMP4 (rabbit, Abcam, Cambridge, UK, ab39973, 1:1000, RT), CD44 (rabbit, Abcam, ab51037, 1:5000, RT), CMYC (rabbit, Abcam, ab32072, 1:1000, RT), CYCLIN D1 (rabbit, Abcam, ab24249, 1:8000, RT), SURVIVIN (mouse, Santa Cruz, Santa Cruz, USA, sc-17779, 1:75, RT), SFRP1 (rabbit, Cell Signaling, Danvers, USA, D5A7, 1:250, RT), ß-AKTIN (mouse, Sigma-Aldrich, A5441, 1:10,000, RT) and SDHA (rabbit, Abcam, ab66484, 1:300, 4 °C). HRP-conjugated secondary antibody (goat anti-mouse or donkey anti-rabbit, both Dianova/Jackson ImmunoResearch Laboratories, Baltimore, USA) was added for 40 min at room temperature. Luminescence detection was performed using Immobilon Western Chemiluminescent HRP Substrate (Millipore) according to the manufacturer’s instructions with Fusion FX7 (Vilber-Lourmat, Eberhardzell, Germany). Cell lysates of the following cell lines were included as positive controls: SFRP1 and BMP4: SK-MEL-28; AXIN2; CYCLIN D1 and SURVIVIN: SW480; CD44: HCV29; CMYC: HeLa.
fixed with 500 µl ice-cold acetone for 20 min at −20 °C. After thawing, PBS with 0.1 % Triton X was added for 5 min. Then rabbit anti-SFRP1 (Cell Signaling, D5A7, 1:100) or mouse anti-E-Cadherin (BD Biosciences, no. 610182, 1:100) antibody was added for 1 h at room temperature. After washing with PBS + 0.1 % Tween 20, fluorescence dye-coupled secondary antibody (goat antirabbit, DyLight™549-conjugated or sheep anti-mouse, DyLight™488-conjugated, both Jackson ImmunoResearch Laboratories, Baltimore, USA) was incubated for 30 min at room temperature in the dark. To visualize nuclei, cells were incubated with DAPI (Invitrogen, Karlsruhe, Germany 1:1000 dilution with PBS) for 5 min at room temperature. Immunofluorescence was evaluated using Olympus BX60 fluorescence microscope (Olympus Europe, Hamburg, Germany).
Immunofluorescence
Activity of Wnt-signaling pathway was investigated using TOPglow/FOPglow TCF Reporter Kit (Millipore). pGL4.10 [lus2] vector (Promega) and pGL4.13 [luc2/SV40] vector (Promega) were used as negative and positive control for firefly luciferase, respectively. pGL4.74 [hRluc/TK] vector (Promega) was cotransfected with TOP- or FOP-plasmid
To verify SFRP1 protein expression in cell lines, immunofluorescence analysis was performed. Cells were cultured in BD Falcon™ 8-well culture slides (BD Biosciences, Heidelberg, Germany) to a confluence of 70–80 % and then
Viability and proliferation assay To investigate functional consequences of SFRP1 overexpression and knockdown, effects on viability and proliferation were analyzed. Therefore, 15,000 cells per well were seeded into white (viability) or clear (proliferation) 96-well plates in RPMI medium. Viability and proliferation were measured after 24, 48 and 72 h with CellTiter-Glo Luminescent Cell Viability Assay (Promega) and QIA58 BrdU Cell Proliferation Assay (Merck), respectively, according to the manufacturer’s protocol using the Multiplate Reader Synergy 2 (BioTek). Wound‑healing assay To analyze effects on migration, wound-healing-assay was performed using Culture-Inserts for Live Cell Analysis (Ibidi, Martinsried, Germany) and photo documentation with Olympus IX81 (Olympus Europe). Transfected and control cells were seeded in culture inserts with a concentration of 300,000 cells/ml (RT112) and 700,000 cell/ml (BFTC905) using 70 µl of cell suspension per well. After cells had grown to a dense cell layer, inserts were removed and growth pattern was documented photographically within 24 h. Area of overgrown surface between transfected cells and controls was compared using Axio Vision Rel 4.8.2 Software (Olympus Europe). TOP flash/FOP flash assay
13
to measure transfection efficiency. RT112 and BFTC905 cells were transfected with 500 ng TOP-, FOP- or controlplasmid and cotransfected with 50 ng Renilla-plasmid using Amaxa Cell Line Nucleofector Kit V (Lonza, Basel, Switzerland) and Nucleofector II (Amaxa Biosystems, Basel, Switzerland) according to the manufacturer’s instructions. Luminescence signal was measured using Dual-Luciferase Reporter Assay System (Promega) in Mithras LB 940 (Berthold Technologies, Wildbad, Germany) multiplate reader. Wnt-signaling activity was calculated with the following formula: TOPnormalized = TOP/Renilla (wild-type and transgenic clone), FOPnormalized = FOP/Renilla (wild-type and transgenic clone), and Ratio = TOPnormalized/FOPnormalized. Ratios >2 were considered as active Wnt-signaling pathway. Statistical analysis For statistical analysis, PASW/SPSS 19.0 (IBM, Armonk, NY, USA) was used. To determine statistical significance of differences in functional cell culture experiments, nonparametrical Kruskal–Wallis test (for k-independent random samples, univariate ANOVA) was used. To determine SFRP1-dependant survival, Kaplan–Meier analysis was performed using log-rank test. Survival probability and survival risk were determined with univariate Cox regression analysis (95 % CI). To correlate patient data among each other and to get significant associations, bivariate correlation with Spearman test and Chi-square test was used. P values
ORIGINAL ARTICLE – CANCER RESEARCH
Functional analyses and prognostic significance of SFRP1 expression in bladder cancer Anja Rogler · Emil Kendziorra · Johannes Giedl · Christine Stoehr · Helge Taubert · Peter J. Goebell · Bernd Wullich · Michael Stöckle · Jan Lehmann · Sabrina Petsch · Arndt Hartmann · Robert Stoehr
Received: 14 January 2015 / Accepted: 16 February 2015 © Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose We previously showed that the Wnt-signaling antagonist SFRP1 (secreted frizzled-related protein 1) is a promising marker in bladder cancer. The aim of this study was to validate the prognostic role and analyze the functional significance of SFRP1. Methods Four bladder cancer cell lines (RT112, RT4, J82 and BFTC905) and one urothelial cell line (UROtsa) were used for functional characterization of SFRP1 expression. Effects on viability, proliferation and wound healing were investigated, and canonical Wnt-pathway activity as well as
A. Rogler · J. Giedl · C. Stoehr · A. Hartmann · R. Stoehr (*) Institute of Pathology, University Hospital Erlangen, FriedrichAlexander-University Erlangen-Nuremberg, Krankenhausstr. 8‑10, 91054 Erlangen, Germany e-mail: robert.stoehr@uk‑erlangen.de E. Kendziorra Clinical Research Unit 179, Department of General, Visceral and Pediatric Surgery, University Hospital Göttingen, Robert‑Koch‑Str. 40, 37075 Göttingen, Germany H. Taubert · P. J. Goebell · B. Wullich Department of Urology, University Hospital Erlangen, FriedrichAlexander-University Erlangen-Nuremberg, Krankenhausstr. 12, 91054 Erlangen, Germany M. Stöckle Clinic for Urology and Children’s Urology, University Hospital Saarland, Kirrberger Straße, 66421 Homburg, Saar, Germany
Wnt-signaling target gene expression was analyzed. Additionally, tissue micro-arrays from two different bladder tumor cohorts were evaluated for SFRP1 expression, and associations with survival and histopathological parameters were analyzed. Results The cell lines RT112, RT4, J82 and UROtsa showed SFRP1 expression. In BFTC905, SFRP1 expression was inhibited by promoter hypermethylation. Wntpathway activity was absent in all cell lines and independent from SFRP1 expression. RT112 and BFTC905 were used for further functional characterization. SFRP1 overexpression resulted in decreased viability and migration in BFTC905 cells. Knockdown of SFRP1 expression in RT112 cells resulted only in marginal effects. In bladder tumors, SFRP1 expression was associated with lower tumor grade, but not with progression in patients with papillary bladder cancer. SFRP1 expressing papillary bladder cancer tumors also demonstrated a tendency to longer overall survival. Conclusions SFRP1 is reducing malignant potential of BFTC905 cells, but not by regulation of canonical Wntsignaling pathway. Other pathways, like non-canonical Wnt or the MAPK pathway, could be activated via SFRP1expression loss. In bladder tumors, SFRP1 has the potential to predict outcome for a subset of papillary bladder tumors. Keywords Bladder cancer · Canonical Wnt signaling · SFRP1 · TOP flash/FOP flash assay
J. Lehmann Urology Practice Prüner Gang, Prüner Gang 15, 24105 Kiel, Germany
Introduction
S. Petsch Tumor Zentrum, Friedrich-Alexander-University ErlangenNuremberg, Carl‑Thiersch‑Str. 7, 91052 Erlangen, Germany
Bladder cancer was the sixth most common cancer in Europe in 2008 with urothelial cell carcinoma (UCC) representing the predominant form in 90 % of all cases (Ferlay et al.
13
2010). With regard to histopathology, clinical behavior and molecular profile, two major groups of UCC can be distinguished (Wu 2005). Noninvasive papillary low-grade tumors are thought to develop from urothelial hyperplasia and are characterized by chromosome 9p/q-deletions as well as by activating mutations in FGFR3, PIK3CA and HRAS. They tend to recur frequently, but have a relative low progression rate (Knowles 2006, 2008). Invasive bladder cancers are thought to arise via flat dysplasia and carcinoma in situ and are characterized by initiating P53 alterations, followed by extensive genomic instability with multiple chromosomal aberrations resulting in loss of tumor suppressor genes like PTEN and RB (Knowles 2006, 2008). Thus, invasive bladder tumors show a high tendency to progress and to metastasize, require aggressive treatments and are still associated with poor 5-year survival rates (Feifer et al. 2011). To provide the best appropriate treatment for every patient and to predict individual progression risk and outcome, new reliable molecular markers are needed. In previous experiments, we could find an association of chromosome 8p deletions with papillary, muscle-invasive growth pattern in 99 bladder tumors. Further analysis identified the candidate gene secreted frizzled-related protein 1 (SFRP1), which was deleted in 38 % of all bladder cancer cases and which was to some extent a consequence of promoter hypermethylation. This expression loss was associated with inferior outcome in patients with papillary, muscle-invasive bladder cancer (Stoehr et al. 2004). SFRP1 expression loss was reported in many other malignancies (e.g., in colorectal, breast or kidney cancer) and correlated with promoter hypermethylation (Dahl et al. 2007; Tanaka et al. 2008; Veeck et al. 2006). Thus, we hypothesized that SFRP1 might act as a tumor suppressor gene in bladder cancer and that it has the potential to predict disease progression in papillary bladder cancer. SFRP1 (NCBI Gene ID 6422) is located at chromosome 8p12-11.1 and is transcribed into a secreted 35 kDa glycoprotein. SFRP1 can act as Wnt-signaling inhibitor, via binding of extracellular Wnt-ligands or direct binding of transmembranous receptor FRIZZLED (FZD) (Finch et al. 1997). The aim of this project was to validate the role of SFRP1 as potential progression marker in bladder cancer using cell culture experiments on the one hand and tumor samples on the other. We sought to clarify functional effects of SFRP1 expression in cell lines using viability, proliferation and wound-scratch assays. To investigate consequences on Wntsignaling activity, we performed TOP flash/FOP flash assay and Wnt-target gene expression analysis of AXIN2, BMP4, CD44, CMYC, CYCLIN D1 and SURVIVIN. Furthermore, SFRP1 expression was investigated immunohistochemically in two different bladder cancer cohorts with 86 and 243 patients, respectively, and correlated with survival data.
13
J Cancer Res Clin Oncol
Materials and methods Patient cohorts and tumor specimen Tissue micro-arrays (TMAs) of two different bladder cancer patient cohorts were used for immunohistochemical analysis of SFRP1: group 1 consisted of 86 patients with non-muscle-invasive (pTa or pT1) papillary tumors and group 2 of 243 patients with advanced bladder tumors (≥pT3 and/or pN1), who all underwent radical cystectomy and received adjuvant chemotherapy. TMAs of the advanced tumor group were constructed at the Institute of Pathology Erlangen after collecting the paraffin blocks from a previous prospective adjuvant chemotherapy study of advanced bladder cancers (Lehmann et al. 2005) originally consisting of 327 patients. Due to tissue availability, only a subgroup of 243 patients (74 %) of the initial cohort could be analyzed. From all blocks, HE sections were made and all histopathological parameters were reviewed by one pathologist experienced in uro-pathology. For this study, IRB approval was obtained from the German Association of Urological Oncology (AUO) as well as informed written consent was obtained from all patients of participating local centers and clinics. All relevant patient characteristics and clinico- and histopathological parameters were summarized previously (Lehmann et al. 2005). Papillary bladder tumors were newly collected for this study from the tumor bank of the Comprehensive Cancer Center Erlangen-EMN located at the Institute of Pathology in Erlangen. Formalin-fixed and paraffin-embedded tumor tissues and corresponding hematoxylin–eosin stained sections were selected, and tumor areas were marked and reevaluated according to histopathological stage and grade by two experienced surgical pathologists (AH, JG). The patient cohort was aged between 27 and 97 years (mean age: 70.3 ± 11.4 years) and consisted of 64 men and 22 women. Clinical follow-up data for the papillary tumor group were obtained in collaboration with the clinical cancer registry (Tumorzentrum, TUZ) of Erlangen. Informed written consent was obtained from all patients of the papillary tumor group, and IRB approval was obtained from local ethics committee of the University Clinic Erlangen for retrospective use of patient material in the context of the Comprehensive Cancer Center tissue bank. All relevant patient characteristics, histopathological data and follow-up are shown in Table 1. Cell lines and transfections For functional analysis of SFRP1-expression, the bladder cancer cell lines RT112, RT4, J82 and BFTC905 as well as the normal urothelial cell line UROtsa were screened
J Cancer Res Clin Oncol Table 1 Patient characteristics of bladder cancer cohorts
Patients Age
Gender
Papillary bladder tumor cohort
Advanced bladder tumor cohort
n = 86 Mean: 70.3 years Median: 72 years (±11.4 years) Range: 29–97 years
n = 243 Mean: 62.1 years Median: 63 years (±8.3 years) Range: 38–81 years
Female: n = 22
Female: n = 56
Male: n = 64
Male: n = 182 n.a.: n = 5
Stage
PUNLMP n = 1
pT1, pN+ n = 5
pTa n = 47
pT2, pN+ n = 28
pT1 n = 32
pT3 n = 144
pT2 n = 4
pT4 n = 40
pT3 n = 1
n.a. n = 26
pT4 n = 1 Grade
Follow-up OS
Follow-up DSS
lg n = 40
G2, hg n = 27
hg n = 45
G3, hg n = 209
PUNLMP n = 1
n.a. n = 7
Alive n = 66
Alive n = 130
Dead n = 15 n.a. n = 5
Dead n = 80 n.a. = 34
Alive n = 70
Alive n = 144
Dead n = 8
Dead n = 66
n.a. n = 8
n.a. = 43
using conventional PCR, qRT-PCR, Western blot analysis and immunofluorescence, respectively (Hatina et al. 2008; Masters et al. 1986; Rieger et al. 1995; Rossi et al. 2001; Tzeng et al. 1996). The malignant melanoma cell line SKMel-28 (Carey et al. 1976) was used as positive control for SFRP1-protein detection, and colorectal carcinoma cell line SW480 (Leibovitz et al. 1976) was used as positive control in TOP flash/FOP flash assay. Cells were cultured in RPMI medium supplemented with 10 % fetal calf serum (FCS), 1 % sodium-pyruvate and 1 % l-glutamine at 37 °C and 5 % CO2. Transfections were carried out in 6-well plates seeding 100,000 and 300,000 cells per well for RT112 and BFTC905, respectively. After 24 h of cell adhesion, SFRP1 was stable overexpressed in BFTC905 using the expression plasmid for human SFRP1 in pCMV6-Neo vector (Origene Technologies, Rockville, USA) and FuGene HD (Roche, Penzberg, Germany) with a ratio of 2:6 (DNA:FuGene) according to the manufacturer’s instructions. For clonal selection, 250 µg/µl G-418 (Neomycin, Roche) was added to the culture medium. After 36 h of cell adhesion, SFRP1 was stable knocked down in RT112 using the anti-SFRP1 shRNA overexpression plasmid in pGFP-V-RS shRNA vector (Origene
Technologies) and TurboFectin 8.0 (Roche) with a ratio of 1:3 (DNA/TurboFectin) according to the manufacturer’s instructions. For clonal selection, 3 µg/ml puromycin (InvivoGen, San Diego, USA) was added to the culture medium. To reverse hypermethylation and acetylation, 200,000 cells were cultured in 3 ml RPMI medium, and 2 µM 5-Aza-2′deoxycytidine (Sigma-Aldrich, Taufkirchen, Germany), 2 µM suberoylanilide hydroxamic acid (SAHA, Biozol, Eching, Germany) and 1.65 µM trichostatin A/vorinostat (Sigma-Aldrich) were added. Isolation of DNA and RNA and cDNA synthesis To investigate SFRP1 promoter methylation in papillary urothelial carcinomas and cell lines, tumor specimens were manually micro-dissected, cells were pelleted and DNA was isolated using the High Pure PCR Template Preparation Kit (Roche) according to the manufacturer’s protocol. To analyze SFRP1 and Wnt-target gene expression with nested and qRT-PCR, respectively, RNA was isolated using RNeasy® Mini Kit (Qiagen, Hilden, Germany) and cDNA was converted using the RevertAid™ H Minus First Strand cDNA Synthesis Kit (Fermentas Life Sciences, St. Leon-Rot, Germany) according to the manufacturer’s instructions. For cDNA synthesis, 1 µg total RNA was used. DNA- and RNA-quality was controlled using the Multiplate Reader Synergy 2 (BioTek, Bad Friedrichshall, Germany). Nested PCR Genetic region of SFRP1 was amplified by nested PCR using two pairs of primers (outer sense: 5′-CAC AACGTGGGCTACAAGAA-3′, outer antisense: 5′-GAAG TGGTGGCTGAGGTTGT-3′, inner sense: 5′-CTACTGG CCCGAGATGCTTA-3′, inner antisense: 5′-GCTGGCAC AGAGATGTTCAA-3′) obtained from Metabion (Martinsried, Germany) in a total volume of 25 µl containing approx. 100 ng DNA, 0.2 mM dNTP (Promega, Mannheim, Germany), 0.18 µM primers, 5 µl 5× buffer (Promega) and 0.0025 U/µl GoTaq (Promega). For the nested PCR, two PCRs were carried out using the outer primer pair and additional 5 % DMSO in the first reaction and using the inner primer pair and 2 µl PCR product from the first reaction as template in the second PCR. The thermal cycling conditions for both reactions were as follows: initial denaturation for 3 min at 95 °C, 35 cycles of denaturation at 95 °C for 1 min, annealing at 62 °C (first PCR) or 60 °C (second PCR) for 1 min, elongation at 72 °C for 1 min and final primer extension at 72 °C for 10 min. GAPDH was used as endogenous control, and for its sufficient amplification only the first PCR was necessary using the following primers (Metabion):
13
J Cancer Res Clin Oncol
sense: 5′-TGGTCACCAGGGCTGCTT-3′ and antisense: 5′-GTCTTCTGGGTGGCAGTGAT-3′. Finally, 15 µl of SFRP1 PCR product and 5 µl of GAPDH PCR product were loaded onto 2.5 % agarose gel. The gel was stained for 20 min with ethidiumbromide and visualized under UV-illumination.
at 60 °C. Data analysis was performed with 7500 Software v2.0.5 (Applied Biosystems), and gene expression ratios were calculated with ΔΔCT-method (Schmittgen and Livak 2008).
Bisulfite treatment and methylation‑specific PCR
Immunohistochemistry was performed on formalin-fixed, paraffin-embedded (FFPE) 4 µm TMA sections of tumor tissue specimens transferred to glass slides. Before using TMAs for expression analysis, whole-bladder tumor sections were stained to avoid a bias of the results by a heterogeneous SFRP1 expression in bladder tumors. SFRP1positive tumors showed a diffuse but not a heterogeneous expression throughout the whole tumor (data not shown). TMAs were stained with rabbit anti-SFRP1 antibody (custom-made, Pineda, 1:600) as described below. Briefly, after deparaffinization, endogenous peroxidase was blocked by incubating in 3 % H2O2 for 10 min followed by a 1-min incubation step with citric acid buffer (pH 6.0) at 120 °C to break crosslinks between proteins. Slides were incubated overnight at room temperature with SFRP1 primary antibody. This was followed by incubation with secondary goat anti-rabbit antibody (1:200, Vector Laboratories, Burlingame, USA) for 30 min at room temperature. Then, slides were incubated for 20 min with ABC solution (antibody–biotin complex VECTASTAIN® Elite ABC kit, Vector Laboratories), followed by 10-min incubation with TSAsolution (TSA™ indirect, Perkin Elmer, Waltham, Massachusetts) and 20-min incubation with ABC according to the manufacturer’s protocols. AEC solution (AEC Peroxidase Substrate Kit, Vector Laboratories) was added until staining intensity was sufficient (approx. 10 min). Slides were counterstained for 2 min with hemalaun (Carl Roth, Karlsruhe, Germany) and mounted with Aquatex (Merck, Darmstadt, Germany). Stainings were examined and evaluated by two experienced uropathologist (JG, AH), and immunoreactivity (IRS = immune reactive score) was scored as follows: Intensity (0 = negative, 1 = weak, 2 = moderate, 3 = strong) and number of tumor cells (in percent) were determined. Number of stained cells was correlated with numbers from 0 to 4. No staining of cells was evaluated as 0, 1–25 % as 1, 26–50 % as 2, 51–75 % as 3 and 76–100 % as 4. Numbers were multiplied with staining intensity, and immunoreactive values between 0 and 12 were created. For SFRP1 staining, two immunoreactive groups were made: group 1 = IRS 0 and 1, group 2 = IRS 2–12.
According to the manufacturer’s protocol, 1 µg DNA was used for bisulfite treatment using the EpiTect® Bisulfite Kit (Qiagen); 2 µl bisulfite-treated DNA was used in two methylation-specific PCRs (MSP): one with primers for methylated SFRP1 promoter (“M-PCR”: sense 5′-TGT AGTTTTCGGAGTTA GTGTCGCGC-3, antisense 5′-CCT ACGATCGAAAACGACGCGAACG-3′, Metabion) and one with primers that detected unmethylated SFRP1-promoter (“UM-PCR”: sense: 5′-GTTTTGTAGTTTTTGGAG TTAGTGTTGTGT-3′, antisense: 5′-CTCAACCTACAA TC AAAAACAACACAAACA-3′, Metabion). PCRs were carried out using 0.16 mM dNTPs, 1.2 µM Primermix, 5 % DMSO, 0.0025 U/µl GoTaq and 5 µl 5× buffer in a total volume of 25 µl. In the “unmethylated PCR,” additionally 1 mM MgCl2 was added. The thermal cycling conditions were as follows: initial denaturation for 5 min at 95 °C, 37 cycles of denaturation at 94 °C for 30 s, annealing at 66.5 °C (“M-PCR”) or 63.8 °C (UM-PCR) for 30 s, elongation at 72 °C for 30 s (“M-PCR”) or 45 s (“UM-PCR”) and final primer extension at 72 °C for 10 min. In the “UMPCR,” GoTaq was added after initial denaturation step; 5 µl of each PCR product was loaded onto 2.5 % agarose gel, stained with ethidiumbromide and visualized under UV-illumination. qRT‑PCR To verify SFRP1 wild-type mRNA expression in cell lines and to control knockdown and overexpression of SFRP1 in RT112 and BFTC905, SYBR Green-based quantitative realtime PCR (qRT-PCR) was performed in 7500 Fast (96-well) Real-time PCR system (Applied Biosystems, Darmstadt, Germany) using 25 ng cDNA template, 100nM SFRP1Primermix (sense: 5′-CTACTGGCCCGAGATGCTTA-3′, antisense: 5′-GCTGGCACAGAGATGTTCAA-3′), 250nM GAPDH-Primermix (sense: 5′-TGGTCACCAGGGCTG CTT-3′, antisense: 5′-AGCTTCCCGTTCTCAGCC-3′) and 6.25 µl SYBR Green Mix (2x) in a total volume of 12.5 µl. From SK-Mel-28 cDNA (SFRP1 positive control), 6.25 ng was used. Thermal cycling conditions were as follows: 2 min 50 °C, 10 min 95 °C, 40 cycles of denaturation at 95 °C for 15 s and combined annealing and elongation at 60 °C for 30 s. PCR was completed with melt curve generation: 15 s at 95 °C, 1 min at 60 °C, 30 s at 95 °C and 15 s
13
Immunohistochemistry
Western blotting To analyze SFRP1 and Wnt-target-protein expression in cell lines, we performed Western blot. Therefore, we isolated whole protein fraction from adherent cells incubating
J Cancer Res Clin Oncol
protein lysis buffer (5 M NaCl, 1 M Tris/HCl pH7.2, 1× Triton X, 0.05 % SDS, 0.5 M EDTA pH 8.0, 2 M DTT) on ice for 15 min in the culture flask and cell buffer suspension for 45 min at 4–8 °C. This was followed by ultrasound sonification for 20 s (1 s on/1 s off, 30 % intensity) with HTU SONI 130 (Heinemann, Gmünd, Germany) and centrifugation for 30 min at 4 °C at 13,200g. Protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific, Bonn, Germany) according to the manufacturer’s protocol. SDS–PAGE was performed with 10 % PAA-gels and 30 µg total protein. 1× Loading buffer (5×: 0.313 M Tris HCl pH 6.8, 10 % SDS (w/v), 0.05 % (w/v) bromphenol blue and 50 % (v/v) glycerol) and 1× DTT (from 20× = 2 M DTT) were added, and lysates were denatured for 5 min at 95 °C. Polyacrylamide gels were run for 15 min at 100 V and then for 60 min at 150 V. Proteins were blotted on nitrocellulose membrane using wet blotting method with Mini Protean® Tetra System (BioRad Laboratories, Munich, Germany) according to the manufacturer’s protocol (25 V, overnight). Membrane was washed with TBS/Tween and blocked for 40 min with BSA (Carl Roth, Karlsruhe, Germany, for AXIN2, BMP4, CD44, CMYC, SFRP1 and SDHA), nonfat dry milk (Carl Roth, for SURVIVIN and ß-AKTIN) or Immunoblot Blocking Reagent (Millipore, Billerica, USA for CYCLIN D1), followed by 1-h incubation with primary antibodies against AXIN2 (rabbit, BioMol/Thermo Fisher Scientific, Bonn, Germany, MA5-15015, 1:133, RT), BMP4 (rabbit, Abcam, Cambridge, UK, ab39973, 1:1000, RT), CD44 (rabbit, Abcam, ab51037, 1:5000, RT), CMYC (rabbit, Abcam, ab32072, 1:1000, RT), CYCLIN D1 (rabbit, Abcam, ab24249, 1:8000, RT), SURVIVIN (mouse, Santa Cruz, Santa Cruz, USA, sc-17779, 1:75, RT), SFRP1 (rabbit, Cell Signaling, Danvers, USA, D5A7, 1:250, RT), ß-AKTIN (mouse, Sigma-Aldrich, A5441, 1:10,000, RT) and SDHA (rabbit, Abcam, ab66484, 1:300, 4 °C). HRP-conjugated secondary antibody (goat anti-mouse or donkey anti-rabbit, both Dianova/Jackson ImmunoResearch Laboratories, Baltimore, USA) was added for 40 min at room temperature. Luminescence detection was performed using Immobilon Western Chemiluminescent HRP Substrate (Millipore) according to the manufacturer’s instructions with Fusion FX7 (Vilber-Lourmat, Eberhardzell, Germany). Cell lysates of the following cell lines were included as positive controls: SFRP1 and BMP4: SK-MEL-28; AXIN2; CYCLIN D1 and SURVIVIN: SW480; CD44: HCV29; CMYC: HeLa.
fixed with 500 µl ice-cold acetone for 20 min at −20 °C. After thawing, PBS with 0.1 % Triton X was added for 5 min. Then rabbit anti-SFRP1 (Cell Signaling, D5A7, 1:100) or mouse anti-E-Cadherin (BD Biosciences, no. 610182, 1:100) antibody was added for 1 h at room temperature. After washing with PBS + 0.1 % Tween 20, fluorescence dye-coupled secondary antibody (goat antirabbit, DyLight™549-conjugated or sheep anti-mouse, DyLight™488-conjugated, both Jackson ImmunoResearch Laboratories, Baltimore, USA) was incubated for 30 min at room temperature in the dark. To visualize nuclei, cells were incubated with DAPI (Invitrogen, Karlsruhe, Germany 1:1000 dilution with PBS) for 5 min at room temperature. Immunofluorescence was evaluated using Olympus BX60 fluorescence microscope (Olympus Europe, Hamburg, Germany).
Immunofluorescence
Activity of Wnt-signaling pathway was investigated using TOPglow/FOPglow TCF Reporter Kit (Millipore). pGL4.10 [lus2] vector (Promega) and pGL4.13 [luc2/SV40] vector (Promega) were used as negative and positive control for firefly luciferase, respectively. pGL4.74 [hRluc/TK] vector (Promega) was cotransfected with TOP- or FOP-plasmid
To verify SFRP1 protein expression in cell lines, immunofluorescence analysis was performed. Cells were cultured in BD Falcon™ 8-well culture slides (BD Biosciences, Heidelberg, Germany) to a confluence of 70–80 % and then
Viability and proliferation assay To investigate functional consequences of SFRP1 overexpression and knockdown, effects on viability and proliferation were analyzed. Therefore, 15,000 cells per well were seeded into white (viability) or clear (proliferation) 96-well plates in RPMI medium. Viability and proliferation were measured after 24, 48 and 72 h with CellTiter-Glo Luminescent Cell Viability Assay (Promega) and QIA58 BrdU Cell Proliferation Assay (Merck), respectively, according to the manufacturer’s protocol using the Multiplate Reader Synergy 2 (BioTek). Wound‑healing assay To analyze effects on migration, wound-healing-assay was performed using Culture-Inserts for Live Cell Analysis (Ibidi, Martinsried, Germany) and photo documentation with Olympus IX81 (Olympus Europe). Transfected and control cells were seeded in culture inserts with a concentration of 300,000 cells/ml (RT112) and 700,000 cell/ml (BFTC905) using 70 µl of cell suspension per well. After cells had grown to a dense cell layer, inserts were removed and growth pattern was documented photographically within 24 h. Area of overgrown surface between transfected cells and controls was compared using Axio Vision Rel 4.8.2 Software (Olympus Europe). TOP flash/FOP flash assay
13
to measure transfection efficiency. RT112 and BFTC905 cells were transfected with 500 ng TOP-, FOP- or controlplasmid and cotransfected with 50 ng Renilla-plasmid using Amaxa Cell Line Nucleofector Kit V (Lonza, Basel, Switzerland) and Nucleofector II (Amaxa Biosystems, Basel, Switzerland) according to the manufacturer’s instructions. Luminescence signal was measured using Dual-Luciferase Reporter Assay System (Promega) in Mithras LB 940 (Berthold Technologies, Wildbad, Germany) multiplate reader. Wnt-signaling activity was calculated with the following formula: TOPnormalized = TOP/Renilla (wild-type and transgenic clone), FOPnormalized = FOP/Renilla (wild-type and transgenic clone), and Ratio = TOPnormalized/FOPnormalized. Ratios >2 were considered as active Wnt-signaling pathway. Statistical analysis For statistical analysis, PASW/SPSS 19.0 (IBM, Armonk, NY, USA) was used. To determine statistical significance of differences in functional cell culture experiments, nonparametrical Kruskal–Wallis test (for k-independent random samples, univariate ANOVA) was used. To determine SFRP1-dependant survival, Kaplan–Meier analysis was performed using log-rank test. Survival probability and survival risk were determined with univariate Cox regression analysis (95 % CI). To correlate patient data among each other and to get significant associations, bivariate correlation with Spearman test and Chi-square test was used. P values
