Background
Prostate cancer is one of the most common malignancies and remains the second leading cause of cancer death in men [
1]. The improved understanding of prostate cancer biology in recent years led to the development of drugs directed against precise tumorigenesis-associated molecular pathways [
2]. Angiogenesis, the development of new blood vessels, is recognized as one of the hallmarks of malignancy and prostate vasculature has been shown to play an important role in regulating the size and function of prostate malignancies [
3,
4]. Accordingly, several anti-angiogenic drugs have been tested in phase II and III trials in prostate cancer patients [
5], including the oral non-selective tyrosine kinase inhibitors Sunitinib and Sorafenib. These two drugs share their activity on vascular endothelial growth factor-receptors (VEGFRs), platelet derived growth factor-receptor beta (PDGFRβ), cKIT and RET, expressed on the cell membrane. In addition, Sorafenib is also able to directly act on the RAF intracellular pathway [
6].
A bulk of evidence indicates that tumor blood vessels differ significantly from normal vessels for the structural organization and for the properties of endothelium [
7‐
12]. This suggests that tumor vascularization depends on mechanisms alternative to the simple recruitment from adjacent tissue of pre-existing blood vessels [
13]. The most remarkable abnormality reported for tumor-derived endothelial cells (TEC) is the chromosomal instability [
9]. In addition, serial analysis of gene expression showed that TEC express genes not shared by blood vessels that reside in normal tissues [
11]. Embryonic genes are expressed also by the endothelial cells derived from tumors [
10,
14,
15]. Finally, TEC present functional alterations linked to increased survival, proliferation and angiogenic properties [
13], as well as resistance to chemotherapeutics [
16]. All these molecular and functional alterations in TEC may result in altered sensitivity to the anti-angiogenic therapy. However, information on the phenotype of TEC derived from prostate tumor and on their sensitivity to anti-angiogenic drugs is limited [
17].
In the present study, we isolated and characterized three lines of TEC from three different prostate cancer human samples (PTEC). Moreover, we evaluated the effect of two anti-angiogenic drugs, Sunitinib and Sorafenib, on typical aspects of the angiogenic process such as the ability to form functional blood vessels in vivo, in vitro proliferation, survival, tubulogenesis and motility. Finally, as androgen receptor (AR) stimulation was reported to promote endothelial cell proliferation, we explored the possible effect of a combined treatment with anti-androgen and anti-angiogenic drugs.
Methods
Cancer tissue sampling
Prostate tissue samples (prostate adenocarcinoma) were obtained by Prof. Arnauld Villers and Prof Xavier Leroy (Dept. of Urology, Regional University Hospital of Lille, France) from 3 patients with a mean age of 58 years (ranging from 57 to 59) who underwent radical prostatectomy. Immediately after prostate removal (delay < 10 min), small pieces of tissue (at least 6 tissue samples from 0.5 to 1 cc) were grossly dissected by the pathologist from the left area, the right peripheral zone and the transitional area. To ensure tissue was malignant and to confirm the Gleason score, histological analysis of sections was performed on each sample by the same pathologist (Table
1). Patient verbal and written information and signed consent form required by the tissue collection unit by law was performed and obtained for all patients. This study was in accordance with the ethical requirements of the tissue collection unit of the Centre Hospitalier Régional Universitaire de Lille, University Lille Nord de France.
Table 1
Characteristics of the patients used for the isolation of PTEC
01
| 57 | 9 (4 + 5) | pT3b N0 M0 | 7.88 | NO |
02
| 59 | 9 (4 + 5) | pT3a N0 M0 | 11.46 | NO |
03
| 58 | 7 (3 + 4) | pT3 N0 M0 | 8.03 | NO |
Drugs
Sunitinib malate (Sigma-Aldrich, St Louis, MO, USA), was resuspended in DMSO to a final concentration of 10 mM and stored at +4°C, according to the manufacturer’s instructions. Sorafenib (Bayer Pharmaceuticals, Leverkusen, Germany) was resuspended in DMSO to a stock concentration of 10 mM and stored at -20°C. Bicalutamide (Casodex) (Sigma-Aldrich, St Louis, MO, USA), was resuspended in DMSO to a stock concentration of 10 mM, according to the manufacturer’s instructions. Drugs were diluted into the culture medium shortly before performing the assays.
Isolation of PTEC and other cell types
Prostate tumor endothelial cells (PTEC) were isolated on the basis of endothelial-specific culture conditions. For the isolation of PTEC, specimens were finely minced and digested in RPMI (Lonza, Basel, Switzerland) containing Collagenase IV (Sigma-Aldrich, St Louis, MO, USA) for 30 minutes at 37°C. After washings in medium plus 10% fetal calf serum (FCS, Seromed, Poly-Labo), the cell suspension was forced through a graded series of meshes to separate the cell components from stroma and aggregates. Cells (2×10
4/cm
2) where then plated in ECAF (Endothelial Cells Attachment Factor, Sigma-Aldrich, St Louis, MO, USA)-coated plates in EndoGRO MV-VEGF medium (Merck-Millipore, Billerica, Massachusetts, USA) containing 5% FCS, and maintained in culture for at least 6 passages. To avoid a possible fibroblast contamination, cells were cultured at passage one for three days with D-valine-substituted DMEM (Sigma-Aldrich, St Louis, MO, USA). Breast tumor endothelial cells (BTEC) were isolated and characterized as previously described [
18]. Human umbilical vein endothelial cells (HUVEC) and microvascular endothelial cells (HMEC) were obtained from the umbilical vein or from derma, respectively, as previously described [
19]. All endothelial cells were maintained in culture in EndoGRO MV-VEGF medium containing 5% FCS.
Human Embryonic Kidney (HEK) 293 and Lymph Node Carcinoma of Prostate (LNCaP) C4- 2 cells were grown in DMEM and RPMI 1684 (Invitrogen), respectively, supplemented with 10% FCS, L-glutamine (5 mM) (Sigma-Aldrich, St Louis, MO, USA) and kanamycin (100 mg/ml) (Sigma-Aldrich, St Louis, MO, USA). Cells were transfected with 2 μg of pcDNA4-AR construct using FuGENE HD reagent (Roche Diagnostics, France), as described [
20].
Flow cytofluorimetric and immunofluorescence analysis
For cytofluorimetric analysis, PTEC lines were detached from plates with a non-enzymatic cell dissociation solution (Sigma-Aldrich, St Louis, MO, USA), washed and stained (30 min at 4°C) with the following fluorescein isothiocyanate (FITC)-, phycoerythrin (PE)-, or allophtcocyanin (APC)-conjugated antibodies: PDGFRβ, CD31 (all from BD Bioscience, Franklin Lakes, NJ, USA) CD105, VEGR2 (all from MiltenyiBiotec, Bergisch Gladbach, Germany), c-KIT (Dako, Glostrup, Denmark), TIE2, VEGFR1, VEGFR3 (all from R&D Systems, Minneapolis, MN, USA). Isotypes (all from MiltenyiBiotec, Bergisch Gladbach, Germany) were used as negative controls. Cells were subjected to cytofluorimetric analysis (FACScan Becton Dickinson, Franklin Lakes, NJ, USA) at each culture passage. Indirect immunofluorescence was performed on cells cultured on chamber slides (Nunc, Roskilde, Denmark). Cells were fixed in 3.5% paraformaldehyde containing 2% sucrose and permeabilized with Hepes-Triton X-100 0.1% for 10 minutes at 4°C. The anti-pan-cytokeratin polyclonal Ab (Biomeda, Foster City, California, USA) was used. Texas Red goat anti-rabbit IgG (Molecular Probes, Eugene, OR, USA) was used as secondary antibody. Hoechst 33258 dye (Sigma-Aldrich, St Louis, MO, USA) was added for nuclear staining. Confocal microscopy analysis was performed using a Zeiss LSM 5 Pascal model confocal microscope (Carl Zeiss, Oberkochen, Germany).
In vitro formation of capillary-like structures was studied on growth factor-reduced Matrigel (BD Bioscience, Franklin Lakes, NJ, USA) in 24-well plates. PTEC, HUVEC and HMEC (3,5 × 104 cells/well) were seeded onto Matrigel-coating in EndoGRO MV-VEGF medium containing 5% FCS and treatments performed in duplicate. Cell organization onto Matrigel was periodically imaged with a Nikon Eclipse Ti inverted microscope using a Nikon Plan 10X/0,10 objective and cells were kept on a stage incubator at 37°C and 5% CO2 during the experiment (OKOLab, Italy). Images were acquired at 2 h time intervals using MetaMorph software.
Image analysis was performed with ImageJ software: images at 18 hours of treatment were analyzed: number of nodes (intersections formed by at least three detectable cells) and total tubule length (aligned cells connecting nodes) were measured for each field. Number of nodes and tubule length were normalized to maximum values and their sum for each condition was used to express the grade of organization in "capillary-like" structures in terms of arbitrary units (A.U.). At least ten fields for each condition were analyzed in each independent experiment. Graphs show mean values of three independent experiments, error bars show standard error. Values are expressed as mean ± S.E.M.
To evaluate the tubule formation
in vivo, 2 × 10
6 PTEC were implanted subcutaneously into SCID mice (Charles River, Wilmington, Massachusetts) within growth factor–reduced Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) as previously described [
10]. Briefly, cells were harvested and resuspended in 150 μl DMEM plus 250 μl of Matrigel, chilled on ice and injected subcutaneously into the left back of SCID mice (n = 4). After 7 days, mice were sacrificed, and endothelial plugs recovered and processed for histology. Typically, the overlying skin was removed, and gels were cut out by retaining the peritoneal lining for support, fixed in 10% buffered formalin, and embedded in paraffin. Sections (3 μm) were cut and stained with hematoxylin and eosin and were examined under a light microscope system.
Proliferation assay and MTT
For the cytotoxicity or the proliferation assay, cells were plated in the growth medium at a concentration of 3000 cells/well in a 96-multiwell plate and left in adhesion overnight. The day after the culture medium was removed and cells were incubated with Sunitinib or Sorafenib in RPMI 2% FCS. After 48 h, DNA synthesis was detected after as incorporation of 5-bromo-2-deoxyuridine (BrdU) using an enzyme-linked immunosorbent assay kit (Roche, Penzberg, Germany). Cytotoxicity was evaluated by MTT (3-(4,5-dimetiltiazol-2-yl)-2,5-difenil tetrasodium bromide) (Merck-Millipore, Billerica, Massachusetts, USA), according to manufacturer’s instructions. Data are expressed as the mean ± S.E.M of the media of absorbance of at least three different experiments performed with all the three lines in the study in triplicate, normalized to the positive control (vehicle alone). To evaluate the IC50 of both Sunitinib and Sorafenib on HUVEC, MTT data were analysed using Calcusyn software. Results are expressed as mean ± S.E.M. of five different experiments.
Cell migration
Migration was assessed using silicone culture inserts (ibidi GmbH, Munich, Germany) in 12-well culture plates. Inserts had two 70 μl wells, both of which were used to plate cells. PTEC, HUVEC or BTEC (1 × 105cells/ml) were plated on 1% gelatin coating in EndoGRO MV-VEGF medium containing 5% FCS. Cells were maintained in incubator until confluence was reached. Cell monolayers were starved 12 hours in DMEM 0% FCS before removing the inserts and thus generating the "wound area". Floating cells were removed by wash in PBS solution, and monolayers were treated with test conditions (in duplicate). EndoGRO MV-VEGF medium 5% FCS was used as positive control, whereas DMEM 0% FCS served as negative control. Cell migration was imaged with a Nikon Eclipse Ti inverted microscope using a Nikon Plan 4X/0, objective and cells were kept on a stage incubator at 37°C and 5% CO2 during the experiment (OKOLab, Italy). Images were acquired at 2 h time intervals using MetaMorph software.
MetaMorph software was used to calculate migration rate (%) by measuring the distance covered by cells between two subsequent time points (4 fields measurements for each image). Measurements were made for each time point and at least 10 fields for each condition were analyzed in each independent experiment. Graphs show mean ± S.E.M of three independent experiments.
Western blot analysis
PTEC and HUVEC were grown in ENDOGRO MV-VEGF medium containing 5 and 10% FCS respectively, until cells reached confluence. Cells were then incubated 10 or 30’ with vehicle, Sunitinib or Sorafenib (1 and 2.5 μM) and lysed at 4°C for 30 min in RIPA buffer (20 nMTrisHCl, 150 nMNaCl, 1% deoxycholate, 0.1% SDS, 1% Triton X-100, pH 7.8) supplemented with protease inhibitor cocktail, PMSF (all from Sigma-Aldrich, St Louis, MO, USA) and PhosStop (Roche, Penzberg, Germany). Conditions for Western blotting were as described previously [
21]. Polyvinylidene fluoride membranes were blocked and incubated overnight with goat polyclonal anti VEGFR2 (R&D Systems, Minneapolis, MN, USA) antibody (1:2000); or rabbit polyclonal anti Phospho-VEGFR2-Tyr951 (p-VEGFR2; sc-101821, Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibody (1:100); rabbit polyclonal anti p44/42MAPK (ERK1/2) (9102, Cell Signalling, Danvers, MA, USA) antibody (1:1000); rabbit polyclonal anti Phospho-p44/42 MAPK (pERK1/2) (Thr202/Tyr204) (4370, Cell Signalling, Danvers, MA, USA) antibody (1:1000); rabbit polyclonal anti Akt (9272, Cell Signalling) antibody (1:3000); rabbit monoclonal IgG Phospho-Akt (pAkt; Ser473) (4508, Cell Signalling) antibody (1:2000); rabbit polyclonal anti AR (N-20) antibody (1:400) (sc-816, Santa Cruz Biotechnology). Membranes were then washed with 1X TBST containing 0.1% Tween 20 and incubated as required with the HRP-conjugated anti-goat (Dako, Glostrup, Denmark) or anti-rabbit IgG (Santa Cruz Biotechnology) antibodies. Chemiluminescence detection was conducted using the ECL prime Western blotting detection reagent (GE Healthcare, Buckinghamshire, England). To quantify the differences in protein phosphorylation, the ratio between non phosphorylated and phosphorylated protein expression was evaluated. Membranes were then washed with 1X TBST containing 0.1% Tween 20 and incubated as required with the HRP-conjugated anti-goat (Dako, Glostrup, Denmark) or anti-rabbit IgG (Santa Cruz Biotechnology) antibodies.
RNA isolation and real time PCR
Total RNA was isolated using Trizol Reagent (Ambion, Life Technologies, Carlsbad, California, USA) according to the manufacturer’s protocol, and quantified spectrophotometrically (Nanodrop ND-1000). For gene expression analysis, quantitative real-time PCR was performed. Briefly, first-strand cDNA was produced from 200 ng of total RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, California, USA).
Quantitative Real-time PCR experiments were performed in 20-μl reaction mixture containing 5 ng of cDNA template, the sequence-specific oligonucleotide primers (purchased from MWG-Biotech, Gmbh, Eurofins Genomics, Hamburg, Germany) and the Power SYBR Green PCR Master Mix (Applied Biosystems). 18S was used to normalize RNA inputs. Fold change expression respect to HMEC was calculated for all samples. The sequence-specific oligonucleotide primers used are: AR (NM_000044.3) forward, 5′-GCAGGAAGCAGTATCCGAAG-3′ (position 1709); reverse, 5′-CTCTCGCCTTCTAGCCCTTT-3′ (position 2067); 18S ribosomal RNA (18S, X03205) forward, 5′- CAGCTTCCGGGAAACCAAAGTC-3′ (position 1132); reverse, 5′- AATTAAGCCGCAGGCTCCACTC -3′ (position 1222). Primer amplification efficiencies were 100.2% for AR and 101.3% for 18S, and the slope values were -3.316 for AR and -3.291 for 18 s respectively. The comparative Ct method was adopted for relative quantification of gene expression and 18 s was used to normalize RNA inputs. Fold change expression respect to HMEC was calculated for all samples.
Statistical analysis
Data are presented as means ± S.E.M. Statistical and significant differences were determined using one-way ANOVA with Newmann-Keuls or Dunnett multicomparison tests (GraphPad Prism version 4.00, GraphPad Software, San Diego, CA) or nonparametric unpaired Wilcoxon-Mann–Whitney test, as appropriate. A p value of < 0.05 was considered significant. Kolmogorov-Smirnov statistical analysis was used to test significant differences in cytofluorimetric data.
Discussion
Taken together, the results of this study show a different sensitivity of endothelial cells isolated from prostate tumors to the anti-angiogenic drugs Sunitinib and Sorafenib. Whereas normal endothelial cells showed similar responses to both drugs in term of proliferation, survival and motility, PTEC were affected by Sunitinib whereas they were more resistant to Sorafenib. However, combined treatment with the anti-androgen Casodex was able to enhance the susceptibility of PTEC to Sorafenib likely trough inhibition of the Akt intracellular pathway.
Several evidences of the literature showed that TEC present in different tumors, including prostate carcinoma, are different from normal endothelial cells at genetic, epigenetic and functional levels [
9,
17,
30,
31]. In particular, recent studies of transcriptome and methylome analysis of endothelial cells from healthy or patients affected by prostate cancer showed a wide spectrum of differences in gene expression and methylation patterns in endothelial cells between malignant and normal prostate tissues [
30,
31]. In addition, murine endothelial cells from spontaneous prostate tumors were reported to display a mesenchymal differentiative ability [
17]. In the present study we isolated and cultured endothelial cells from prostate tumors from patients without androgenic ablation therapy. PTEC were able to migrate, organize in capillary-like structures
in vitro and in vessel structures
in vivo, connected with the mouse vasculature, indicating their endothelial phenotype. As previously reported [
26], PTEC expressed higher AR levels than normal endothelial cells indicating the persistence of the phenotype of origin. The ability of TEC to organize into functional vessels
in vivo has been previously described to be characteristic of TEC isolated from human tumors, at variance with HUVEC that undergo apoptosis [
10,
17,
18].
PTEC may therefore represent a suitable model to assess the response to anti-angiogenic drugs and the related cell signal mechanisms. Indeed, although the results of anti-angiogenic therapy in preclinical models of prostate cancer provided promising results, some discrepancy between these data and those obtained in clinical trials were observed [
32]. In particular, monotherapy treatment of patients with advanced prostate cancer partially failed the endpoints [
27,
33‐
36]. A phase III study comparing Sunitinib versus placebo showed a progression free survival but not an overall survival improvement in Sunitinib treated patients, although phase II studies showed a PSA decline in plasma [
33,
37]. Phase II studies with Sorafenib showed a regression of metastases but not PSA decline [
34,
36]. Is therefore evident that additional knowledge on endothelial characteristics in prostate cancer is required.
In the present study, we showed that both drugs had a cytotoxic effect on normal endothelial cells with a similar IC50 at 48 hour around 1.5 μM. This is in line with previous observations showing that both Sorafenib and Sunitinib are well known inhibitors of pro-angiogenic functions in normal endothelial cells [
16,
38]. Accordingly, Sunitinib induced a dose dependent reduction of proliferation and survival in PTEC as well as in HUVEC and HMEC. In contrast, Sorafenib only partially affected PTEC proliferation and survival. Both drugs slightly reduced cell motility, with a consistent lower effect of Sorafenib. In addition, Sunitinib, but not Sorafenib, affected tubulogenesis in both HUVEC and PTEC. As Sunitinib and Sorafenib share common targets [
5,
6], the differential effect of these drugs on PTEC appears unexpected and of interest.
Studies comparing the in vitro activity of Sorafenib and Sunitib on endothelial cells are limited in the literature. A similar effect of the drugs on cell viability was previously reported in neuroblastoma and corneal epithelial cells [
39,
40]. At variance, our results, showing that PTEC were less sensible to Sorafenib than to Sunitinib, are in line with the reported resistance to Sorafenib of endothelial cell from hepatocellular carcinoma [
16].
As PTEC only expressed the VEGFRs, and not other surface targets of Sunitinib and Sorafenib, such as PDGFRβ and cKIT, we reasoned that the differential response to these drugs could depend on a differential effect on VEGFR2 phosphorylation [
6]. However, VEGFR2 phosphorylation was inhibited by both drugs, excluding this possibility. On the other hand, these drugs were reported to affect different intracellular pathways, including the Ras/Raf/MEK/ERK, JAK/STAT and the PI3K/AKT pathways [
27‐
29]. It is conceivable that the increased resistance to Sorafenib observed in PTEC may be due to its activity on intracellular pathways differentially activated in normal and tumor endothelial cells, as reported [
10,
41]. In PTEC, we observed that Sorafenib and Sunitinib treatments differentially modulated Akt phosphorylation, as no inhibitory effect of Sorafenib was observed on Akt activation. These data correlate with the functional resistance to the effect of Sorafenib observed on PTEC behavior
in vitro. On the contrary, no major effect of these drugs was observed on the p44/42MAPK (ERK1/2) pathway. In this regard, Kharazhia and co-workers recently described an increased Sorafenib resistance of highly-metastatic compared with non-metastatic prostate cancer cells which was due to constitutively active PI3K/AKT pathway targeted by Sorafenib [
29].
Due to the important role of ARs in sustaining prostate cancer progression, second line hormone therapy is frequently employed in prostate cancer to target persistent AR activation. In addition, AR has been described to play a role in the regulation of endothelial cell proliferation [
26]. Moreover, the association of Sorafenib with anti-androgen therapy (Casodex) in a recent phase II clinical trial induced PSA decline and stable disease [
42], improving the effect of the anti-angiogenic monotherapy. Accordingly, our results combining Sorafenib and Casodex successfully overcame the PTEC resistance to Sorafenib both at the functional level and on the Akt pathway activation. Therefore, it can be inferred that the resistance to Sorafenib treatment involves the Akt pathway; which is in turn affected by the combined treatment with Casodex.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AFP, NP and BB designed the research; AV and XL carried out the prostatectomy and collected patients’ data; AB isolated the cells, carried out flow cytometry, proliferation and cytotoxicity assays, protein and RNA isolation, Real Time PCR; BB and AB carried out immunoflorescence analysis and in vivo tubule formation. MB and AFP carried out in vitro tubule formation and cell migration experiments; TG and GG performed Western blotting; MB carried out the statistical analysis; BB, AFP, AB, MB and DG wrote the paper. All of the authors have been involved in revising the manuscript and have given final approval of the version to be published.