S1P、LPA在癌细胞中表达情况

RESEARCH ARTICLE Expression of sphingosine-1-phosphate receptors and lysophosphatidic acid receptors on cultured and xenografted human colon, breast, melanoma, and lung tumor cells Reinhard Müller & Christoph Berliner & Jessica Leptin & Daniel Pörtner & Wojciech Bialecki & Burkhard Kleuser & Udo Schumacher & Novica M. Milićević Received: 28 December 2009 /Accepted: 21 April 2010 /Published online: 18 May 2010 # International Society of Oncology and BioMarkers (ISOBM) 2010 Abstract The lysophospholipids sphingosine-1-phosphate (S1P) and lysophosphatidic acid (LPA) are small lipid molecules with a variety of physiological roles. Addition- ally, their involvement in the initiation and progression of malignant tumors has been increasingly recognized in recent years. However, the data on the expression of S1P and LPA receptors on different cancer cells are very few. Real-time polymerase chain reaction was used for the analysis of mRNA expression of five S1P(1–5) and three LPA(1–3) receptors on a large panel of 13 colon, breast, melanoma, and lung cancer cell lines. Furthermore, the modulation of S1P and LPA receptor mRNA expression was studied upon xenotransplantation of tumor cells into severe combined immunodeficient (SCID) mice. The S1P and LPA receptors were expressed to a variable degree on all tumor cell lines tested (with exception of colon cancer SW480). Most notably, tumor cell lines in vitro expressed S1P2 mRNA that was down-regulated upon xenotransplan- tation, whereas LPA2 receptor mRNA was strongly expressed both in vitro and in vivo (except by breast cancer cells). The latter was especially distinctive for small cell lung tumor cells. The S1P and LPA receptors are differentially expressed on tumor cell lines in vitro. Their expression is modulated upon xenografting into SCID mice in vivo. Keywords Sphingosine-1-phosphate . Lysophosphatidic acid . Receptor . mRNA . Cancer . SCID Introduction The lysophospholipids sphingosine-1-phosphate (S1P) and lysophosphatidic acid (LPA) are small bioactive lipids which exert a variety of roles in normal physiology. S1P and LPA activate diverse groups of G-protein-coupled receptors: there are five S1P receptors (S1P1–5) and three LPA receptors (LPA1–3). These receptors are widely expressed within normal tissues and regulate decisive cellular functions, such as cellular Ca2+ homoeostasis and cytoskeletal reorganization, proliferation and survival, migration and adhesion [1]. Thus, S1P and LPA have been implicated in development, as well as in regulation of essential functions within the cardiovascular, immune, and nervous system [2]. In addition to their physiological roles, it has been shown that S1P and LPA signaling is also involved in the regulation of pathophysiological processes, including auto- immune and immunodeficiency diseases, arteriosclerosis, and cancer. In particular, S1P and LPA receptor signaling has been related to the development and progression of malignant tumors [3,4]. R. Müller :C. Berliner : J. Leptin :D. Pörtner :W. Bialecki : U. Schumacher Center for Experimental Medicine, Institute of Anatomy II: Experimental Morphology, University Hospital Hamburg-Eppendorf, Hamburg, Germany B. Kleuser Institute of Nutritional Science, University of Potsdam, Potsdam, Germany N. M. Milićević (*) Institute of Histology and Embryology, Faculty of Medicine, University of Beograd, Višegradska 26, 11000 Belgrade, Serbia e-mail: emilicen@etf.bg.ac.rs Tumor Biol. (2010) 31:341–349 DOI 10.1007/s13277-010-0043-7 The elevated expression of one or more of S1P receptors has been shown to occur in various human malignant cells. Generally, the expression of S1P1 or S1P3 has been related with stimulation and of S1P2 with the inhibition of migratory behavior of thyroid [5], gastric [6] and breast cancer cells in culture [7]. In addition to human tumor cells, similar results were obtained with B16 mouse melanoma cell line and related with the modulation of its metastatic potential: overexpression of S1P1 or S1P3 led to stimulation of cell migration and aggravation of metastasis, whereas the overexpression of S1P2 inhibited cell migration and metastasis [8,9]. However, these experiments were per- formed only in vitro [5–8] or by the intravenous injection of tumor cells [9], which resulted in dissemination rather than in true metastasis—hence, it is questionable whether this presumably simplistic view of the S1P receptor action truly depicts the biological behavior of malignant cells. On the other hand, aberrations in production and degradation of LPA have been detected, e.g., markedly elevated levels of LPA were observed in the ascitic fluid of ovarian cancer patients [4,10]. Cell culture studies indicated that autocrine LPA activation loops exist in ovarian and prostate cancer cells [11,12]. Thus, in combination with altered LPA receptor expression, LPA may represent an important regulator in the pathophysiology of cancer [10]. However, the data on the expression of various LPA receptors on tumors other than ovarian and prostate cancers are very sparse in the literature. Therefore, we examined the expression of S1P and LPA receptors using a large panel of different cancer cell lines on the one hand and, on the other, investigated the modulation of their expression upon xenotransplantation into severe combined immunodeficient (SCID) mice. Here, we show that S1P and LPA receptors are differentially expressed in tumor cell lines in vitro and that their expression is modulated upon xenografting into SCID mice in vivo. Material and methods Cell lines The human melanoma cell lines MV3 and MeWo cell lines were kindly provided by the Klinik für Dermatologie, Universitätsklinikum Hamburg-Eppendorf, Germany. FEMX-1 cell line was kindly provided by Prof. Øystein Fodstad (University in Oslo, Norway). The small cell lung cancer cell lines H69, SW2, OH1, OH3, and H82 were kindly provided by Prof. Uwe Zangemeister-Wittke (Uni- versity of Bern, Switzerland). Breast cancer MCF7, HBL100, and colon cancer HT29, H29mdr, SW480 cell lines were obtained from the European Collection of Animal Cell Cultures (Porton Down, Wiltshire, UK). The brief characteristics of cell lines are shown in Table 1. Mycoplasma screening was routinely performed for all cell lines and only mycoplasma-free cells were used. Human cancer cell lines were maintained routinely with RPMI 1640 medium (with L-glutamine and 2 g of glucose/L) containing 10% fetal calf serum (FCS; Invitrogen, Karlsruhe, Germany), 100 U/ml penicillin and 10,000 μg/ml strepto- mycin in humidified tissue culture incubator at 37°C, in 5% CO2 atmosphere. At subconfluency, adherent cells were washed once with phosphate-buffered saline (PBS; 50 mM phosphate, 150 mM NaCl; pH 7.4) and trypsinized. The reaction was stopped with RPMI 1640 medium and the cells were collected (centrifugation at 1,500 rpm for 3 min) as a cell pellet prior to lysis and total RNA isolation. Small cell lung cancer cells growing in suspension were resus- pended in medium and collected (centrifugation at 1,500 rpm for 3 min) as a cell pellet prior to lysis and total RNA isolation. Paraffin-embedded tumors For injection into SCID mice, viable human cancer cells (5×106) were suspended in 1 ml of cell culture medium. A sample of 200 μl of this suspension was injected subcuta- neously between the scapulae of each SCID mouse. The mice bearing tumors were sacrificed when the tumor had reached maximal growth (20% of the body weight of the animal at the beginning of the experiment) or started to ulcerate. The tumors were removed, weighed and fixed in 10% neutral buffered formalin and processed to paraffin wax (for further details see [13,14]). Mice SCID mice were obtained from our in-house animal facility. Animals housed up to five to a cage were provided with sterile food and water ad libitum. In addition to the tumor experiments, three SCID mice, aged 12–14 months, were killed in accordance with animal welfare guidelines and 200–800 μl of blood per mouse were collected in EDTA tubes. RNA isolation Total RNA from cell lines Cancer cells (15–20×106 per sample) were resuspended in a 2-ml lysis buffer from the RNeasy® Midi Kit (Qiagen, Hilden, Germany). The cell suspension was disrupted and homogenized by passing the lysate at least five times through a needle (0.9 mm diameter) fitted to a syringe. The sample was mixed with 2 ml of 70% ethanol and 342 Tumor Biol. (2010) 31:341–349 transferred to an RNeasy spin column. All washing steps followed the manufacturer’s instructions. Total RNA was eluted in 350 μl of RNase-free water. Total RNA from paraffin-embedded tumors Using a scalpel, excess paraffin was trimmed off the sample block. Eight 10-μm sections per sample were pooled and immediately placed in a 2-ml microcentrifuge tube. All preparation steps followed the manufacturer’s instructions for the RNeasy® FFPE Kit (Qiagen, Hilden, Germany). The total RNA was eluted in 30 μl of RNase-free water. Total RNA from mouse blood Total RNAwas isolated from 200 μl mouse blood per sample following the manufacturer’s instruction for the QIAamp® RNA Blood Mini Kit (Qiagen, Hilden, Germany). The total RNA was eluted in 50 μl of RNase-free water. All RNA solutions were stored at −20°C prior to cDNA synthesis. We isolated total RNA from fresh cell lines and paraffin- embedded tumors and compared the results. It was found that the source of material did not make any difference. cDNA synthesis The cDNA synthesis was performed in a Biometra thermal cycler (Biometra, Göttingen, Germany) in a total volume of 20 μl for each sample and followed the manufacturer’s instruction for the First Strand Transcriptor cDNA Synthe- sis Kit (Roche Diagnostics, Mannheim, Germany). cDNA synthesis with total RNA from cell lines, with total RNA from paraffin-embedded tumors and with total RNA from mouse blood Anchored-oligo(dT)18 primer were used for reverse transcrip- tion of 1.5 μg of total RNA into cDNA for each sample. Eleven microliters of total RNA solution per sample was reversely transcribed into cDNAwith random hexamer primer. Anchored-oligo(dT)18 primer were used for reverse transcrip- tion of 300 ng of total RNA into cDNA for each sample. Polymerase chain reaction primers All primer pairs were created by the Universal Probe Library (Roche Diagnostics, Mannheim, Germany). All primer pairs were synthesized by MWG-BIOTECH AG (Ebersberg, Germany). The primer forward a was high- purity salt-free and primer forward b and reverse were HPLC purified (see Table 2 for primers). Standard curve and calibrator For each gene, a tenfold serial dilution of its polymerase chain reaction (PCR) product (dilution from 1:10−3 to 1:10−10 in PCR-grade water; amplified with forward a and reverse primer) was used as a template for the external standard curve using the PCR-run protocols indicated above with forward b and reverse primer pairs. A 1:10−7 dilution was used as calibrator in the PCR run amplifying β-ACT and a 1:10−8 dilution was used as calibrator in the PCR run amplifying S1P1-5 and LPA1-3. Cell line Cell line derived from Metastatic in SCID mice Breast cancer MCF7 Pleural effusion from breast cancer patient + HBL 100 Breast milk cells; however, part of SV40 genome integrated in this cell line, so not strictly normal or strictly cancer − Colon cancer HT29 Adenocarcinoma + H29mdr Multi-drug resistance phenotyp of HT29 cells −/+ SW480 Grade 3-4 adenocarcinoma of the colon − Melanoma FEMX-1 Metastastic melanoma lymph node + MV-3 Metastastic melanoma lymph node + MeWo Metastastic melanoma lymph node + Small cell lung cancer OH1 Pleural effusion from 43 year old male + OH3 Pleural effusion from 43 year old male + H69 Pleural effusion from 56 year old male + H82 Pleural effusion from 56 year old male + SW2 Bone marrow aspirate + Table 1 The characteristics of tumor cell lines + metastatic, +/− weakly meta- static, − nonmetastatic Tumor Biol. (2010) 31:341–349 343 Analysis of the specificity of PCR products All primer pairs were tested for human specificity. Therefore, 250 ng of RNA from human cancer cell lines or from SCID mouse peripheral blood lymphocytes were reverse transcribed into cDNA. The amplification of the genes was specified by melting point analysis in the Light Cycler, in an agarose gel electrophoresis, and in sequencing analysis (data not shown). Polymerase chain reaction The template for the standard curve and the calibrator for each gene were amplified in a Biometra thermal cycler (Biometra, Göttingen, Germany). Two microliters of cDNA solution was used as a template for the PCR and incubated in a total reaction volume of 20 μl. The PCR master mix (Qiagen, Hilden, Germany) included Taq DNA polymerase, Taq PCR buffer, a dNTP mixture, MgCl2, and forward a and reverse primer, 10 pmol each. The PCR conditions were denaturation: 30 s 94°C; annealing: 30 s 53°C (S1P1)/56°C (S1P2)/55°C (S1P3)/50°C (S1P4)/52°C (S1P5)/ 52°C (LPA1)/55°C (LPA2)/54°C (LPA3); and elongation: 60 s 72°C. Real-time polymerase chain reaction Real-time polymerase chain reaction and melting curve analyses for the relative quantification of gene expression were performed in 100-μl glass capillaries using the LightCycler Fast Start DNA MasterPLUS SYBRGreen I Kit (Roche Diagnostics GmbH, Mannheim, Germany) and the LightCycler 2.0 System. Analysis of gene expression Ten microliters of cDNA solution (1:10 diluted in PCR- grade H2O) or 10 μl calibrator (1:10 −7 diluted in PCR grade H2O) were used as a template for the PCR reaction and incubated in a total reaction volume of 50 μl (in 100-μl glass capillaries). The SYBR Green I Master mix included Taq DNA polymerase, Taq PCR buffer, a dNTP mixture, 1 mmol/l MgCl2, and 25 pmol (β-actin), 50 pmol (S1P1–5) forward b and reverse primer, respectively. The PCR conditions were initially 10 min at 95°C, followed by 50 cycles of denaturation: 15 s at 95°C, annealing: 15 s at 62°C (for all primer pairs), and elongation: 26 s 72°C. Fluorescence was measured at the end of the elongation phase. Melting curve analysis (0 s 95°C, 12 s 65°C, and 0 s 95°C (0.1°C/s)) was performed directly after each PCR run. The list of all primers is shown in Table 2. Human umbilical cord vein endothelial cells The human umbilical cord vein endothelial cells (HUVECs) were used as control to study the expression of LPA receptors on a non-malignant cell line. The HUVECs were obtained from PromoCell GmbH (Heidelberg, Germany) and for subcultivation DetachKit solution of the same manufacturer was used. The HUVECs were cultured under the identical conditions and the procedures for HUVECs were the same as for the cancer cell lines used in this study. Briefly, RNA was gained from pelleted cells using Qiagen RNeasy Midi Kit (Qiagen GmbH, Hilden, Germany). The protocol was performed as recommended by the manufac- turer, except the elution step. Here, 130 μl RNase-free water was used two times for elution, i.e., the last step was repeated twice. Two independent experiments were per- Table 2 List of primers Primer name Primer sequence Product size (bp) hACT forw a aga aaa tct ggc acc aca cc 190 hACT forward b cca acc gcg aga aga tga 97 hACT rev cca gag gcg tac agg gat ag S1P1 forw a cct ctt gtg ccc tta aaa gc 171 S1P1 forw b attactttaactggtagggaacg 151 S1P1 rev aagacatctctcggtttaattgc S1P2 forw a caa tgt acc tgt ttc tgg gc 225 S1P2 forw b ggccttcgtagccaatacct 173 S1P2 rev tgccatacagcttgaccttg S1P3 forw a cca tta act cta cta ggg agc 193 S1P3 forw b gccaccatttccactaggag 168 S1P3 rev gca tat tgg tgc aca ttg gt S1P4 forw a aaa tgg gct tcc cat ggt cac c 220 S1P4 forw b gagagcaccctggtgtgg 158 S1P4 rev catgatcgaacttcaatgttgc S1P5 forw a act ctg gta tca gaa ccg 219 S1P5 forw b ccacgactgtcttcccaagt 179 S1P5 rev caagcagaacgtcaattcca hACT forw a aga aaa tct ggc acc aca cc 190 hACT forward b cca acc gcg aga aga tga 97 hACT rev cca gag gcg tac agg gat ag LPA1 forw a tagttctggggcgtgttcac 110 LPA1 forw b gcgtgttcaccacctacaac 100 LPA1 rev tgtggttcattcatggctgt LPA2 for A tggcagagcatgtcagctgc 150 LPA2 for b cagcctggtcaagactgttgt 104 LPA2 rev tgcaggactcacagcctaaa LPA3 for A atacaagtgggtccatcagc 163 LPA3 for b acggtgatgactgtcttaggg 113 LPA3 rev caccttttcacatgctgcac 344 Tumor Biol. (2010) 31:341–349 formed in duplicate. The following RNA was collected: HUVECs pellet I with 172.2 ng/μl and HUVECs pellet III with 195.9 ng/μl. HUVECs II and IV were the controls doubling the HUVECs I and III in order to minimize pipetting failures. The protocols for HUVECs II and IV were identical to those used for HUVECs I and III. This RNA was transformed by Transcriptor First Strand cDNA Synthesis Kit (Roche Diagnostics GmbH, Mannheim, Germany) into cDNA using the protocol recommended by the manufacturer. In both experiments, 7.5 μl of RNA was used. LightCycler protocols used in this set of experiments for relative quantification (Roche Diagnostics GmbH, Mannheim, Germany) were identical to those in the experiments with cancer cell lines. The list of primers used in these experiments is shown in Table 2. The negative control contained only water that was used in every probe together with primers and the LightCycler reaction mix (Roche Diagnostics GmbH, Mannheim, Germany) to ensure that no contamination with DNA has occurred. MV3 cells previously amplified in a semi-nested PCR with the forward A primers were used as calibrators at concentration of 10−5. Calibrators 1 and 2 were identical (as a double check). Calibrators 1 and 2 were used to compare the results with the standard curves of LPA1–3. Immunohistochemistry Formalin fixed, Difco Agar Noble embedded and routinely processed to paraffin wax HBL100 cells and HUVECs were used for immunohistochemistry. After deplastination by xylene and a series of graded ethanols, antigen retrieval was achieved by incubating the sections in an 85°C hot water bath in citrate buffer, pH=6.0 for 16 h. After transfer to PBS, non-specific binding sites were blocked by normal swine serum (Dako, Glostrup, Denmark) diluted 1:10 for 30 min at room temperature. The sections were then incubated with the rabbit polyclonal primary antibody against amino acids 322–381 of S1P1 of human origin (Edg-1; H60; Santa Cruz, CA; sc-25489) diluted 1:25 at 4°C overnight or with the rabbit isotype IgG (X0903; Dako, Glostrup, Denmark), as a specificity control. The concentration of the immunoglobulin was adjusted to the concentration of the antibody requiring a dilution of 1:2,500. After three washes in PBS, the sections were incubated with biotinylated swine anti-rabbit secondary antibody (E0353; Dako, Glostrup, Denmark) diluted 1:200 at room temperature for 30 min. The signal was detected using an ABC-AP kit for 30 min at room temperature and sequential addition of Liquid Permanent Red Chromagen (K0640; Dako, Glostrup, Denmark) for 10 min. Slides were counterstained with Hämalaun 1:1 for 10 s and mounted with Aqua Tex (1.08562; Merck, Darmstadt, Germany). Statistics Relative quantification was calculated using LightCycler Software 4.05 (Roche Diagnostics, Mannheim, Germany). ANOVA graphs were calculated and plotted with Graph- Pad software v4.0 (GraphPad Software, Inc., San Diego, CA, USA) and p<0.05 was considered statistically significant. Results Expression of sphingosine-1-phosphate receptor mRNA on cultured tumor cells Our analysis revealed a complex picture of S1P receptor mRNA expression on cultured tumor cell lines. Most cell lines showed notable expression of some S1P receptor mRNA except SW480, MCF7 and MeWo, which did not express any of them (Fig. 1a–e). The other cells generally expressed multiple S1P receptor mRNA, except HT29mdr that expressed only S1P2 receptor mRNA (Fig. 1b). The expression of S1P2 mRNA was most consistent across the different cell lines and it was registered on large majority of cultured tumor cells (Fig. 1b). The expression