Formylpeptide Receptor FPR and the Rapid Growth of Malignant Human Gliomas

Journal of the National Cancer Institute, Vol. 97, No. 11, June 1, 2005 ARTICLES 823

Formylpeptide Receptor FPR and the Rapid Growth of Malignant Human Gliomas Ye Zhou , Xiuwu Bian , Yingying Le , Wanghua Gong , Jinyue Hu , Xia Zhang , Lihua Wang , Pablo Iribarren , Rosalba Salcedo , O. M. Zack Howard , William Farrar , Ji Ming Wang

Background: The formylpeptide receptor (FPR) is a G- protein – coupled receptor (GPCR) that mediates chemotaxis of phagocytic leukocytes induced by bacterial peptide N -formyl -methionyl-leucyl-phenylalanine (fMLF). We previ-ously showed that selected human glioma cell lines also express functional FPR. We therefore investigated the relationship between FPR expression and the biologic behavior of glioma cells. Methods: Expression and function of FPR in the human glioblastoma cell line U-87 were examined by reverse transcription – polymerase chain reaction (RT-PCR) and chemotaxis assays, respectively. FPR protein expression was detected in specimens from 33 human primary gliomas by im-munohistochemistry. FPR short interfering (si) RNA was used to block FPR expression in U-87 cells. Cell proliferation was assessed by measuring DNA synthesis. Xenograft tumor for-mation and growth were measured in nude mice. Endogenous FPR agonist activity released by necrotic tumor cells was assessed by measuring FPR activation in an FPR-transfected basophil leukemia cell line and live U-87 cells. Vascular endothelial growth factor (VEGF) mRNA was assessed by RT-PCR, and VEGF protein was assessed by enzyme-linked immunosorbent assay. All statistical tests were two-sided. Results: FPR was selectively expressed by the highly malig-nant human glioblastoma cell line U-87 and most primary grade IV glioblastomas multiforme and grade III anaplastic astrocytomas. U-87 cells responded to the FPR agonist fMLF by chemotaxis (i.e., increased motility), increased cell prolif-eration, and increased production of VEGF protein. FPR siRNA substantially reduced the tumorigenicity of U-87 cells in nude mice (38 days after implantation, mean tumor volume from wild-type U-87 cells = 842 mm 3 , 95% confi dence interval [CI] = 721 to 963 mm 3 ; and from FPR-siRNA transfected U-87 cells = 225 mm 3 , 95% CI = 194 to 256 mm 3 ; P = .001). Necrotic glioblastoma cells released a factor(s) that activated FPR in live U-87 cells. Conclusions: FPR is expressed by highly malignant human glioma cells and appears to mediate motility, growth, and angiogenesis of human glioblastoma by interacting with host-derived agonists. Thus, FPR may represent a molecular target for the development of novel antiglioma therapeutics. [J Natl Cancer Inst 2005;97:823 – 35]

Chemoattractant receptors are seven transmembrane G- protein – coupled receptors (GPCRs) that mediate cell migration in response to a variety of chemotactic factors. Chemoattractant GPCRs also participate in essential pathophysiologic processes, including infl ammation, hematopoiesis, development, wound healing, human immunodefi ciency virus infection, and, most in-triguingly, in the progression of malignant tumors. In fact, some

GPCRs for chemokines promote angiogenesis, thus contributing to neovascularization and tumor outgrowth ( 1 , 2 ) . In addition, some chemokines interact with their GPCRs to induce chemo-taxis of malignant human and animal tumor cells and direct their organ-specifi c metastasis ( 3 , 4 ) . Consequently, GPCRs that respond to chemoattractants may also increase the motility and dissemination of malignant tumor cells.

Glioma is the most common malignant neoplasm of the central nervous system. These tumors range in degree of aggres-siveness from slowly growing low-grade tumors to rapidly growing high-grade tumors, such as anaplastic astrocytoma and glioblastoma multiforme. High-grade tumors contain necrotic foci and are richly vascularized, presumably as a result of aberrant expression of angiogenic factors, such as vascular endothelial growth factor (VEGF), by tumor cells. Established human glioma cell lines and normal astrocytes express several GPCRs for chemokines and respond in vitro to selected ligands by chemotaxis and intracellular calcium mobilization ( 5 , 6 ) . In addition, we previously reported ( 7 ) that some human glioma cell lines express another GPCR, the formylpeptide receptor (FPR). FPR, originally identifi ed in phagocytic leukocytes, mediates the chemotaxis and activation of these cells in response to the bacterial chemotactic peptide N -formyl-methionyl-leucyl-phenylalanine (fMLF) ( 7 ) and potential agonist peptides derived from mitochondria of ruptured cells. Agonist binding to FPR in phagocytic leukocytes leads to the activation of phosphati-dylinositol 3-kinase (PI3K), mitogen-activated protein kinases (MAPKs), and the transcription factor nuclear factor (NF)- κ B [for review, see ( 1 ) ]. Because FPR was believed to mainly engage in proinfl ammatory and antibacterial host responses ( 8 , 9 ) , the unexpected expression of this GPCR by glioma cells ( 7 ) prompted us, in this study, to investigate the contribution of FPR to glioma cell motility, proliferation, and tumorigenicity and to the progression of tumors in vivo. In addition, we explored whether FPR interacted with agonists derived from necrotic tumor cells.

Affi liations of authors: Laboratory of Molecular Immunoregulation (YZ, YL, JH, PI, OMZH, WF, JMW) and Laboratory of Experimental Immunology (XZ, RS), Center for Cancer Research, and Basic Research Program (WG, LH), SAIC-Frederick, Inc., NCI–Frederick, Frederick, MD; Institute of Pathology, Southwest Hospital, The Third Military Medical University, Chongqing, 400038, China (XB).

Present affi liation: Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China (YL).

Correspondence to: Ji Ming Wang, MD, PhD, LMI, CCR, NCI-Frederick, Building 560, Room 31-40, Frederick, MD 21702 – 1201 (e-mail: [email protected] ).

See “ Notes ” following “ References. ”

DOI: 10.1093/jnci/dji142 Journal of the National Cancer Institute, Vol. 97, No. 11, © Oxford University Press 2005, all rights reserved.

by guest on February 17, 2016http://jnci.oxfordjournals.org/

Dow

nloaded from

http://jnci.oxfordjournals.org/

824 ARTICLES Journal of the National Cancer Institute, Vol. 97, No. 11, June 1, 2005

M ATERIALS AND M ETHODS

Reagents

fMLF was from Sigma-Aldrich (St. Louis, MO). The inhibi-tor of MAPK kinase 1 (MEK1) PD98059, antibodies against phosphorylated extracellular signal-regulated kinases 1/2 (ERK1/2), p38 MAPK, Jun-N-terminal kinase (JNK), Akt, or signal transducers and activators of transcription 3 (STAT3), and antibodies against total ERK1/2, p38, JNK, Akt, STAT3, Bcl-2, or Bcl-xL were from Cell Signaling Technology (Beverly, MA). The tyrosine kinase inhibitor Tyrphostin AG490 was from Calbiochem (San Diego, CA). Anti-hypoxia inducible factor (HIF)-1 α antibody was from Novus Biologicals (Littleton, CO). Anti-VEGF antibody was from R&D Systems (Minneapolis, MN). Anti- β - actin antibody was from Santa Cruz (Santa Cruz, CA). Anti-glial fi brillary acidic protein (GFAP) and anti- vimentin antibodies and the EnVision system were from DAKO (Carpin teria, CA). t - Butyloxycarbonyl-methionyl-leucyl-phenylalanine (tBoc-MLF) was from MP Biomedicals (Irvine, CA). Anti-FPR antibody was from BD Biosciences Pharmingen (San Diego, CA). Cyclosporin H (CsH) was a kind gift from Novartis (Basel, Switzerland).

Cells

Human glioblastoma cell lines U-87 and SNB75 were from the American Type Culture Collection (ATCC, Manassas, VA). SHG-44 cells were established from a surgically removed human astroglioma (Suzhou Medical University, Suzhou, China). These cells were grown in Dulbecco’s modifi ed Eagle medium (DMEM) containing 10% fetal calf serum. A rat baso-phil leukemia cell line transfected with the FPR gene (ETFR cells), from Dr. R. Snyderman (Duke University, Durham, NC), was maintained in DMEM with 10% fetal calf serum and G418 (Invitrogen, Carlsbad, CA) at 0.8 mg/mL. The collection of hu-man umbilical cord vein endothelial cells (HUVECs) was previ-ously described ( 10 ) . Human peripheral blood mononuclear cells were isolated from leukopacks obtained from the Transfusion Medicine Department, National Institutes of Health Clinical Center (Bethesda, MD), by Ficoll-Hypaque (Sigma-Aldrich) density gradient centrifugation. Monocytes were purifi ed from human peripheral blood mononuclear cells by Percoll gradient centrifugation (Amersham Biosciences, Little Chalfont, UK) to yield preparations that were more than 90% pure ( 11 ) .

Immunohistochemistry

Tumor specimens from 33 glioma patients were examined for FPR expression by immunohistochemical staining. All pa-tients received diagnosis and surgical therapy and provided written informed consent at the Southwestern Hospital, Third Military Medical University, Chongqing, China. The histologic diagnoses were based on the 2000 edition of World Health Or-ganization (WHO) classifi cation of nervous system tumors. The tumor specimens studied included 13 grade II astrocytomas, 14 grade III anaplastic astrocytomas, and six grade IV glioblasto-mas multiforme. Tumor specimens were fi xed in 10% formalin and embedded in paraffi n. Sections were incubated with 0.3% H 2 O 2 for 10 minutes, blocked with 10% normal goat serum for 10 minutes at room temperature, and reacted with anti-human

FPR antibody for 2 hours at 37 °C, followed by biotinylated secondary antibody (30 minutes, at room temperature) and streptavidin- peroxidase complex (30 minutes, at room tempera-ture). Tumor sections were stained for antibody against FPR protein by use of diaminobenzidine as substrate to give a brown color and were then counterstained with hematoxylin. To detect GFAP and vimentin, glioma cells that had been cultured on chamber slides were fi xed in ethanol, permeabilized with Triton X-100, and incubated with normal goat serum to block poten-tial nonspecifi c binding of antibodies to the cells. Antibodies against GFAP or vimentin were added, and slides were treated with a horseradish-conjugated secondary antibody for visual-ization of the cellular proteins by use of an EnVision system (DAKO).

RT-PCR

Total RNA was extracted from cells with RNeasy Mini Kit (QIAGEN, Valencia, CA), and 0.5 μ g was used for reverse transcription – polymerase chain reaction (RT-PCR). For human FPR, sense primer 5 ′ -CTCCAGTTGGACTAGCCACA-3 ′ and antisense primer 5 ′ -CCATCACCCAGGGCCCAATG-3 ′ gener-ated a 500-bp product. For human VEGF, primers were de-signed to amplify four isoforms of the VEGF gene. Sense primer 5 ′ -ATGAACTTTCTGCTGTCTTGGG-3 ′ and antisense primer 5 ′ -CTGTATCAGTCTTTCCTGGTGAG-3 ′ generated 514-bp VEGF121, 646-bp VEGF165, 718-bp VEGF189, and 769-bp VEGF206 products ( 12 ) . RT-PCR was performed with a ProSTAR High Fidelity single-tube RT-PCR system (Stratagene, La Jolla, CA), consisting of a 30-minute reverse transcription at 42 °C; a 1-minute inactivation of Moloney murine leukemia virus reverse transcriptase at 95 °C; 40 cycles (30 cycles for VEGF) of denatur-ing at 95 °C for 30 seconds, annealing at 60 °C for 30 seconds, and extension at 68 °C for 2 minutes; and a fi nal 10-minute extension at 68 °C. Primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as controls (Stratagene). PCR products were electrophoresed on 2% agarose gels and visualized with ethidium bromide staining.

Directional Cell Migration (Chemotaxis)

Chemotaxis assays were performed in 48-well chemotaxis chambers (NeuroProbe, Gaithersburg, MD) ( 7 , 10 ) . The upper and lower compartments of the chemotaxis chambers were separated by a 10- μ m (pore-sized) (8 μ m for HUVECs) polycar-bonate fi lter (GE Osmonics Labstore, Minnetonka, MN) coated with collagen type I (BD Biosciences, San Jose, CA) at 50 μ g/mL. A 27- μ L aliquot of chemoattractants was placed in the wells of the lower compartment, and 50 μ L of tumor cells, ETFR cells, or HUVECs (each at 1 × 10 6 cells per mL of RPMI 1640 medium containing 1% bovine serum albumin and 25 m M HEPES) were placed in the wells of the upper compartment. A 270-minute incubation at 37 °C was used to measure tumor and ETFR cell migration, and a 120-minute incubation was used to measure HUVEC migration. After incubation, the fi lters were removed and stained, and cells that migrated across the fi lters were counted under light microscopy. The results were expressed as the means, and 95% confi dence intervals (CIs), of migrated cells in three high-powered fi elds (×400 magnifi cation) in triplicate samples.


Dow

nloaded from



Immunoblot Analysis

Tumor cells were lysed in 150 μ L of 1 × sodium dodecyl sulfate sample buffer (62.5 m M Tris-HCl at pH 6.8, 2% sodium dodecyl sulfate, 10% glycerol, and 50 m M dithiothreitol), sonicated for 3 seconds, and then boiled for 5 minutes to produce a cell lysate. The cell lysate was then centrifuged at 10 000 × g at 4 °C for 10 minutes. Immunoblot analysis of Bcl-2 and Bcl-xL and of phos-phorylated ERK1/2, p38, JNK, Akt, and STAT3 proteins was per-formed with antibodies for the specifi c proteins and phosphorylated proteins, respectively. Total cell proteins were electrophoresed on 4% – 12% gradient Tris-Glycine precast gels (Invitrogen) and transferred onto Immobilon P membranes (Millipore, Billerica, MA). The membranes were blocked by incubation in 3% nonfat dry milk for 3 hours at room temperature and then incubated with primary antibodies in phosphate-buffered saline (PBS) containing 0.01% Tween-20 overnight at 4 °C. After incubation with a horse-radish peroxidase – conjugated secondary antibody, the protein bands were detected with Super Signal Chemiluminescent Sub-strate Stable Peroxide Solution (Pierce, Rockford, IL) and BIOMAX-MR fi lm (Eastman Kodak, Rochester, NY). To detect β -actin, total ERK1/2, p38, JNK, Akt, or STAT3, the membranes were stripped with Restore Western Blot Stripping Buffer (Pierce) and then incubated with the corresponding specifi c antibodies. To detect HIF-1 α , glioblastoma cells were washed with fi ve volumes of hypotonic buffer (10 m M HEPES, 10 m M KCl, 1.5 m M MgCl 2 , and 0.5 m M dithiothreitol at pH 7.9) and lysed in the same buffer supplemented with 1% Nonidet P-40. After centrifugation at 10 000 × g at 4 °C for 1 hour, the nuclei-containing pellet was re-suspended in 150 μ L of low-salt buffer (10 m M HEPES, 25% glycerol, 1.5 m M MgCl 2 , 20 m M KCl, 0.5 m M dithiothreitol, and 0.2 m M EDTA) and 150 μ L of high-salt buffer (low-salt buffer containing 800 m M KCl). The nuclear extracts were then centri-fuged as described above, and supernatants were electrophoresed on a 4% – 12% gradient Tris-Glycine precast gel (Invitrogen). The proteins were transferred to Immobilon P membranes, and the blot was probed with anti-HIF-1 α antibody.

Electrophoretic Mobility Shift Assay

The electrophoretic mobility shift assay was used to detect the formation of complexes between a [ 32 P]dATP-labeled nucleotide element (5 ′ -AGTTGAGGGGACTTTCCAGGC-3 ′ ) and NF- κ B contained in the nuclear proteins of tumor cells. The labeled nucleotide element was incubated with 5 μ g of nuclear proteins in 20 μ L of binding mixture (50 m M Tris-HCl at pH 7.4, 25 m M MgCl 2 , 5 m M dithiothreitol, and 50% glycerol) at 4 °C for 2 hours. For supershift assays, nuclear extracts were incubated with 1 μ g of normal rabbit serum or antiserum specifi c for p50/p65 NF- κ B at 4 °C for 1 hour and then with 32 P-labeled oligonucleotide for 15 minutes at room temperature. Nucleotide – protein complexes were resolved on 5% polyacrylamide gels containing 0.25× TBE (Tris – borate – EDTA) buffer at room temperature for 2 hours at 150 V. The gels were then heat-dried under vacuum and exposed to X-Omat fi lms (Eastman Kodak) at − 70 °C.

VEGF Production and the Formation of Capillary-Like Structures by HUVECs

VEGF in the supernatants of tumor cells was quantifi ed by com-mercial enzyme-linked immunosorbent assay kits (Lymphokine

Testing Laboratory, SAIC Frederick, Frederick, MD). The for-mation of capillary-like structures by HUVECs was measured after culturing the cells on 24-well plates coated with Matrigel (BD Biosciences) at 0.3 mL per well. Matrigel in the wells was solidifi ed at 37 °C for 1 hour, and 4 × 10 4 HUVECs in 0.5 mL of conditioned medium from tumor cells were added to each well. The capillary-like structure formed by HUVECs after 20 hours was photographed at ×200 magnifi cation under a phase – contrast microscope.

Tumor Cell Proliferation

Tumor cell proliferation was assessed by measuring the incor-poration of [ 3 H]thymidine during DNA synthesis. Briefl y, U-87 cells were cultured in 96-well tissue culture plates at 8000 cells per well in DMEM containing 10% fetal calf serum for 12 hours. After a 1-hour treatment with various inhibitors, the cells were incubated in DMEM containing 0.5% fetal calf serum alone or 0.5% fetal calf serum in the presence of 1 μ M fMLF. Cells were then pulse-labeled with 1 μ Ci (in a 5- μ L volume containing 20 nmol) of [ 3 H]thymidine (Amersham Pharmacia Biotech, Piscataway, NJ) per well for 16 hours and harvested onto mem-branes by use of an Inotech Harvester (Inotech Biosystems, Rockville, MD). The incorporated [ 3 H]thymidine was measured with a Wallac Microbeta Counter (Perkin-Elmer Life Sciences, Gaithersburg, MD), and the results were expressed as mean counts per minute (cpm) and 95% confi dence interval of four replicates.

Xenografts

Approximately 5 × 10 6 human glioma cells (in 100 μ L of PBS) were implanted by subcutaneous injection into the fl ank of each 4-week-old (20 – 22 g) female athymic Ncr-nu/nu mouse (NCI-Frederick Cancer Research Facility). Tumor size was calculated by the formula lw 2 /2, where l is the length of the tumor in milli-meters and w is the width in millimeters. Each group contained at least fi ve mice. To generate xenografts with glioblastoma cells transfected with FPR short interfering RNA (siRNA), 1 × 10 6 cells were injected into the fl ank of each nude mouse. Nontrans-fected U-87 cells and cells transfected with random siRNA (mock) were used as controls. Animal care was provided in accordance with the Guide for the Care and Use of Laboratory Animals.

FPR siRNA and Transfection of U-87 Cells

Construction of hairpin siRNA expression cassettes was per-formed as described ( 13 ) . Briefl y, three 19-nucleotide sequences were targeted to FPR mRNA at nucleotides 392 to 410 (in the second transmembrane region of the putative protein), nucleotides 605 to 623 (in the third transmembrane region), and nucleotides 926 to 944 (in the third extracellular loop) (GenBank sequence no. NM_002029 ). Retroviral vector stocks were produced by transient transfection of Phoenix-Ampho cells with the Superfect Transfection Reagent (QIAGEN) and 5 μ g of FPR siRNA ex-pression plasmid. The virus was collected from the culture super-natants on day 2 after transfection, and U-87 cells were transfected with a combination of three retroviral vectors, each containing a separate siRNA construct in the presence of Polybrene at 5 μ g/mL. The U-87 cells stably transfected with FPR siRNA were se-lected and maintained by continual incubation with puromycin (BD Biosciences Clontech, Palo Alto, CA) at 2 μ g/mL.


Dow

nloaded from



Ca 2+ Flux

Ca 2+ mobilization was measured by incubating 2 × 10 7 cells in 1 mL of loading medium (DMEM containing 10% fetal calf serum and 2 m M glutamine) with 7 μ M Fura-2 acetoxymethyl ester (Molecular Probes, Eugene, OR) for 45 minutes at room temperature. The dye-loaded cells were washed and resuspended in saline buffer (138 m M NaCl, 6 m M KCl, 1 m M CaCl 2 , 10 m M HEPES, 5 m M glucose, and 0.1% bovine serum albumin at pH 7.4) at a density of 0.5 × 10 6 cells per mL. The cells were then transferred into quartz cuvettes (1 × 10 6 cells in 2 mL of saline buffer), and cuvettes were placed in a fl uorescence spectrometer (Perkin-Elmer, Beaconsfi eld, UK). Stimulants were added to a cuvette in a volume of 20 μ L, and the intensity of the fl uores-cence was measured by use of the ratio of the absorbance at 340 nm to the absorbance at 380 nm, calculated with an FL WinLab program (Perkin-Elmer).

Generation of Tumor Cell Supernatant and Tissue Extracts

Necrotic U-87 cells were generated by subjecting 20 × 10 6 cells in 1 mL of PBS to three cycles of freezing and thawing followed by centrifugation. Apoptotic U-87 cells were gener-ated by treating 20 × 10 6 cells in 1 mL of PBS with 0.5 μ M staurosporine for 6 hours. Supernatants from necrotic U-87 tumor tissues were obtained by mincing tumor tissues from

xenograft nude mice at a concentration of 1 g of tissue in 4 mL of PBS and then subjecting the mixture to repeated cycles of freezing and thawing followed by centrifugation. For both types of cells, centrifugation was at 10 000 × g for 1 hour (4 °C), and supernatants were collected and stored at − 70 °C for further assays.

Statistical Analyses

All experiments were performed at least three times. A t test, with the computer-aided program Prism (Version IV) (GraphPad Software, Inc. San Diego, CA), was used to determine the statis-tical signifi cance of the difference between cell responses to test-ing materials and to controls in chemotaxis and cell proliferation experiments as well as comparison of tumor volumes. P values equal to or less than .05 were considered statistically signifi cant. All statistical tests were two-sided. Mouse survival curves were plotted as Kaplan – Meier plots (Prism Version IV, GraphicPad Software, Inc.).

R ESULTS

Functional Characteristics of FPR in Glioblastoma Cells

As previously described ( 7 ) , we confi rmed that an established human glioblastoma cell line, U-87 ( 14 ) , expressed FPR tran-scripts ( Fig. 1, A ) and responded to stimulation with the FPR

Fig. 1. Expression of formylpeptide receptor (FPR) and FPR signaling in human glioblastoma cells. A ) Expression of FPR mRNA. The expression of FPR mRNA in normal astroglial cells and in the glioma cell lines U-87 and SHG-44 was examined by reverse transcription – polymerase chain reaction (RT-PCR). Human monocytes were used as controls. Lane 1 = monocytes; lane 2 = normal astrocytes; lane 3 = SHG-44 cells; lane 4 = U-87 cells. RT-PCR product of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as the loading control. B ) Chemotaxis of U-87 cells in response to N -formyl-methionyl-leucyl-phenylalanine (fMLF). Different concentrations of fMLF (as indicated) were placed in the lower wells of the chemotaxis chamber; U-87 cells were placed in the upper wells, which were separated from lower wells by a 10- μ m (pore-size) polycarbonate fi lter coated with collagen type I. After incubation, the cells that migrated across the fi lter were counted under light microscopy. Results are expressed as the mean number of migrated cells in three high-powered (×400)

fi elds. Error bars = 95% confi dence intervals. *, P values range from .024 to <.001, compared with cell migration in response to control medium (two-sided t test). C ) Phosphorylation of extracellular signal-regulated kinases 1/2 (ERK1/2), p38, Jun-N-terminal kinase (JNK), and Akt in U-87 cells stimulated with 10 2 n M fMLF as indicated. Cells were lysed, the lysates were electrophoresed on Tris-Glycine precast gels, and then proteins were transferred to Immobilon P membranes. The blots were probed with specifi c antibodies against phosphorylated proteins, followed by incubation with a horseradish peroxidase – conjugated secondary antibody. The phosphorylated protein bands were detected with a Super Signal Chemiluminescent Substrate Stable Peroxide Solution and Biomax-MR fi lm. Antibodies against total proteins were used to demonstrate the levels of protein loading. Incubation times were as follows: lanes 1 = 0 minute; lanes 2 = 1 minute; lanes 3 = 3 minutes; lanes 4 = 5 minutes; lanes 5 = 10 minutes; lanes 6 = 20 minutes; lanes 7 = 30 minutes; lanes 8 = 60 minutes; lanes 9 = 120 minutes.


Dow

nloaded from



agonist fMLF in nanomolar concentration range by chemotaxis ( Fig. 1, B ). To examine whether FPR transcripts were expressed by all astroglial cells, we used another human glioma cell line, SHG-44, and normal human astroglial cells. Neither SHG-44 nor normal human astrocytes expressed FPR transcripts ( Fig. 1, A ), and these cells did not respond to fMLF at any concentration tested (data not shown), suggesting that functional FPR is ex-pressed by some human glioma cells.

The selective expression of FPR by U-87 cells prompted us to examine FPR signaling pathways by probing for increased phos-phorylation of signaling components that may contribute to the malignant behavior of these tumor cells, including those that may promote tumor cell proliferation and gene transcription. We found that the levels of phosphorylation of ERK1/2, p38 MAPK, and JNK in U-87 glioblastoma cells increased after the cells were stimulated with fMLF ( Fig. 1, C ). Activation of FPR in U-87 cells also promoted the phosphorylation of Akt ( Fig. 1, C ), also known as protein kinase B, which is located downstream of the PI3K pathway and has been reported to support tumor cell survival and proliferation ( 15 ) . Therefore, PI3K, which controls the activation of Akt, and MAPK appear to be coupled to FPR in U-87 cells.

Because activation of FPR in myeloid cells apparently induces nuclear translocation of NF- κ B ( 16 ) , which regulates the transcrip-tion of diverse genes coding for cytokines and growth factors, we investigated whether NF- κ B contained in the nuclear proteins iso-lated from fMLF-stimulated U-87 cells formed complexes with an

NF- κ B – binding nucleotide element. As early as 5 minutes after fMLF stimulation, we detected the formation of complexes that contained the NF- κ B subunits p50 and p65 ( Fig. 2, A ). In addition, in fMLF-stimulated U-87 cells under normal oxygenated culture conditions, we also detected the nuclear translocation of HIF-1 α . HIF-1 α has been reported to increase the transcription of the VEGF gene, followed by increased production of VEGF protein. VEGF protein recruits vascular endothelial cells and thus promotes angio-genesis ( 17 ) . Addition of the MEK1 inhibitor PD98059 inhibited the fMLF-stimulation of HIF-1 α translocation in glioblastoma cells ( Fig. 2, B ), suggesting that the ERK1/2 MAPK pathway is involved in fMLF-induced HIF-1 α translocation in U-87 cells.

The transcription factor STAT3 has also been implicated in increasing angiogenesis in malignant tumors and is coupled to the signaling pathways of some GPCRs for chemoattractants ( 18 , 19 ) . We investigated whether fMLF could activate STAT3 in glioblastoma cells by measuring the level of STAT3 phosphory-lation and found that fMLF induced a rapid and transient phos-phorylation of STAT3 in U-87 cells at both tyrosine-705 (Tyr-705) and serine-727 (Ser-727) residues ( Fig. 2, C ). In fMLF-treated monocytes, transiently increased phosphorylation was observed only at Ser-727 of STAT3. In untreated monocytes, Tyr-705 was constitutively phosphorylated at a relatively high level, which was not further increased after fMLF stimulation ( Fig. 2, D ). The reason for the difference observed in FPR-induced STAT3 acti-vation between glioblastoma cells and monocytes is not clear.

Fig. 2. Activation of transcriptional factors in glioblastoma cells by the formylpeptide receptor (FPR) agonist N -formyl-methionyl-leucyl-phenylalanine (fMLF). A ) Activation of nuclear factor κ B (NF- κ B) in fMLF-stimulated U-87 cells. U-87 cells were incubated with 10 2 n M fMLF at 37 °C as follows: Lane 1 = 0 minute; lane 2 = 5 minutes; lane 3 = 20 minutes; lanes 4 to 8 = 40 minutes. The electrophoretic mobility shift assay was performed to detect the binding of NF- κ B contained in the nuclear proteins of tumor cells to the NF- κ B – binding nucleotide element. Nuclear extracts from U-87 cells were incubated with the 32 P-labeled oligonucleotide NF- κ B – binding element in the presence or absence of normal rabbit serum or antiserum specifi c for p50/p65 NF- κ B. DNA-protein complexes were resolved on polyacrylamide gels, which were then exposed to X-Omat fi lms. Lanes 1 to 4 = nuclear extracts incubated with labeled probe; lane 5 = nuclear extract incubated with both labeled and unlabeled probe; lane 6 = supershift in the presence of anti-p50 antiserum; lane 7 = supershift in the presence of anti-p65 antiserum; lane 8 = normal rabbit serum. B ) Nuclear translocation of hypoxia-inducible factor (HIF)-1 α . U-87 cells were preincubated

in the presence or absence of PD98059 (an inhibitor of MAPK kinase 1 [MEK1]) at 50 μ M , as indicated, for 60 minutes and then treated with 10 2 n M fMLF, as indicated, for 4 or 6 hours at 37 °C. The nuclear proteins were extracted, separated by electrophoresis, and transferred to Immobilon P membranes. The blots were then probed with an anti-HIF-1 α antibody. + = added to incubation mixture; – = not added to incubation mixture. C and D ) Activation of signal transducers and activators of transcription 3 (STAT3). U-87 cells ( C ) or human monocytes ( D ) were incubated with 10 2 n M fMLF, for the times indicated, or were incubated for 20 minutes with the indicated concentration of fMLF. Total cell proteins from lysed cells were electrophoresed and transferred to Immobilon P membranes. The blots were then incubated with primary antibodies against STAT3 phosphorylated at Tyr-705 (p-Tyr705) or at Ser-727 (p-Ser727), followed by incubation with a horseradish peroxidase – conjugated secondary antibody. The protein bands were detected with Super Signal Chemiluminescent Substrate Stable Peroxide Solution and Biomax-MR fi lm. Total STAT3 was detected with an antibody against nonphosphorylated STAT3.


Dow

nloaded from



It should be noted, however, that U-87 cells express approximately 500 high-affi nity binding sites for fMLF per cell ( 7 ) , whereas monocytes express approximately 4500 sites per cell. Whether this could account for the difference in FPR-induced signaling between monocytes and U-87 glioblastoma cells requires further investigation. Thus, FPR in U-87 cells can apparently stimulate the phosphorylation of STAT3, and phosphorylation at both Ser-727 and Tyr-705 has been reported to be required for maxi-mal transcriptional activity of STAT3 ( 20 , 21 ) .

Activation of FPR and VEGF Production by U-87 Cells

Our fi nding that fMLF activated the transcription factors NF- κ B, HIF-1 α , and STAT3 in U-87 cells by activating FPR prompted us to examine whether fMLF could also stimulate the production of VEGF in these cells. Untreated U-87 glioblastoma cells ex-pressed a low level of VEGF mRNA, whereas fMLF-treated cells expressed an increased level of VEGF mRNA ( Fig. 3, A ). fMLF-treated U-87 cells also secreted elevated levels of VEGF protein, which reached a maximum at 72 hours after stimulation with fMLF ( Fig. 3, B ). There was no further increase in the production

of VEGF by U-87 cells stimulated by fMLF for longer periods of time (data not shown).

The VEGF contained in the culture medium of fMLF- stimulated U-87 cells was biologically active, because the condi-tioned medium increased the chemotaxis of HUVECs, and the chemotactic activity in the conditioned medium was completely neutralized by the addition of a monoclonal antibody against human VEGF ( Fig. 3, C ). When HUVECs were cultured with conditioned medium from fMLF-treated U-87 cells, HUVECs formed capillary-like structures on a Matrigel surface, and for-mation of these structures was inhibited by the addition of anti-VEGF antibody ( Fig. 3, D ). Thus, VEGF in conditioned medium from FPR-activated U-87 cells appears to induce endothelial cells to migrate and to form capillary-like structures, two key events associated with neovascularization.

FPR Activation and U-87 Cell Proliferation

We measured DNA synthesis as a refl ection of cell prolifera-tion. When U-87 cells were cultured under suboptimal condi-tions (i.e., culture medium containing only 0.5% fetal calf serum), DNA synthesis in fMLF-treated cells was higher than

Fig. 3. Production of vascular endothelial growth factor (VEGF) by N -formyl-methionyl-leucyl-phenylalanine (fMLF) – stimulated glioblastoma cells. A ) Expres-sion of VEGF mRNA in fMLF-treated U-87 cells. Cells were incubated with fMLF at 0 to 10 3 n M for 6 hours. Total RNA was extracted, and VEGF transcripts were amplifi ed by reverse transcription – polymerase chain reaction (RT-PCR). Various dilutions (1 : 2, lanes 1 ; 1 : 4, lanes 2 ; 1 : 8, lanes 3 ) of the RT-PCR products were electrophoresed on agarose gels and visualized with ethidium bromide staining. The glyceraldehyde-3-phosphate dehydrogenase (GAPDH) RT-PCR product was used as a loading control. B ) Level of VEGF protein in supernatants of U-87 cells treated with 10 2 n M fMLF. Cells were incubated with 10 2 n M fMLF, and culture medium was collected, as indicated, to measure VEGF levels by enzyme-linked immunosorbent assays. The experiments were repeated three times, and each time point consisted of three repli cate samples. Results from a representative experiment are shown. The results are presented as the mean values. Error bars = 95% confi dence intervals (CIs). C ) Chemotaxis. Recombinant human VEGF (5 ng/mL) or supernatants from U-87 cells treated with fMLF (containing VEGF at 5 ng/mL

or 50 ng/mL) were preincubated with control immunoglobulin G (IgG) or anti-VEGF antibody (each at 1 μ g/mL) for 45 minutes at room temperature and then tested for chemotactic activity with human umbilical cord vein endothelial cells (HUVECs). The number of cells in three high-powered fi elds were counted, and data are expressed as the mean. Error bars = 95% CIs. *, P values range from .008 to .002, compared with cells treated with IgG (two-sided t test). Bars 1 = HUVEC migration to control medium; bars 2 = migration to recombinant human VEGF; bars 3 = migration to fMLF-stimulated U-87 cell supernatant containing VEGF at 5 ng/mL; bars 4 = migration to supernatant containing VEGF at 50 ng/mL; bars 5 = migration to 50 ng/mL of the stromal cell-derived factor 1 α (SDF-1 α ) as a positive control. D ) Formation of capillary-like structures on Matrigel. Culture supernatants from U-87 cells treated with 10 2 n M fMLF for 72 hours were preincubated with anti-VEGF antibody (+) or IgG ( – ) (each at 1 μ g/mL) for 45 minutes. HUVECs were then mixed with the supernatants (containing VEGF at 50 ng/mL) and examined for tubule formation on Matrigel after 20 hours. Photomicrographs were made with a phase – contrast microscope. Bar = 50 μ m.


Dow

nloaded from



that in untreated cells (e.g., at 72 hours, [ 3 H]thymidine incorpo-ration in cells cultured in the medium alone = 29 000 cpm [taken as 100%], 95% CI = 25 700 to 32 300 cpm; [ 3 H]thymidine incor-poration in cells cultured with fMLF = 40 000 cpm [a 37% increase compared with the medium-alone group], 95% CI = 36 500 to 43 500 cpm; P <.001) ( Fig. 4, A ). Addition of CsH, a well-characterized FPR inhibitor ( 1 ) , to the cells blocked the effect of fMLF ([ 3 H]thymidine incorporation in cells cultured in medium alone = 25 000 cpm [taken as 100%], 95% CI = 21 200 to 28 800 cpm; [ 3 H]thymidine incorporation in cells cultured in the presence of fMLF = 34 000 cpm [36% increase compared with the medium-alone group], 95% CI = 28 000 to 40 000 cpm; [ 3 H]thymidine incorporation in cells cultured with both fMLF and CsH for 72 hours = 22 000 cpm [89% of the medium-alone group], 95% CI = 18 200 to 25 800 cpm; P = .013 for the CsH+fMLF group versus the fMLF-alone group) ( Fig. 4, B ). Similarly, the tyrosine kinase inhibitor Tyrphostin AG490, which disrupts JAK/STAT signaling ( 22 , 23 ) , also inhibited fMLF-stimulated U-87 cell growth (at 72 hours, medium alone = 31 000 cpm [taken as 100%], 95 % CI = 25 000 to 37 000 cpm; the fMLF group = 50 000 cpm [a 61% increase compared with the medium-alone group], 95% CI = 46 200 to 53 800 cpm; AG490+fMLF group = 28 000 cpm [90% of the medium con-trol], 95% CI = 25 100 to 30 900 cpm; P = .001 between fMLF group and AG490+fMLF group) ( Fig. 4, C ), suggesting that FPR regulates U-87 cell growth by activation of the JAK/STAT cascade.

We next investigated whether fMLF treatment altered the expression of the antiapoptotic molecules Bcl2 and Bcl-xL. We found that fMLF-treated U-87 cells had higher levels of Bcl-2 than untreated cells but had the same levels of Bcl-xL ( Fig. 4, D ). fMLF-activated FPR increased the level of Bcl-2 via the MEK1/ERK pathway, because addition of the MEK1 inhibitor PD98059, but not of the p38 inhibitor SB202190 or JAK/STAT inhibitor Tyrphostin AG490, blocked the fMLF-induced increase in ex-pression of Bcl-2 ( Fig. 4, D , and data not shown).

FPR and Tumorigenesis of Glioma Cell Lines

We next investigated the relationship between FPR and the degree of glioma cell malignancy by examining the expression of vimentin, a marker for poorly differentiated astroglial cells ( 24 ) , and of GFAP, a glial differentiation marker ( 25 ) . We found that U-87 cells contained higher levels of vimentin ( Fig. 5, A ) and lower levels of GFAP than another glioma cell line, SHG-44, which does not express functional FPR ( Fig. 5, A ). These results led us to hypothesize that highly malignant human glioblastomas selectively express FPR.

We then compared tumor growth of U-87 cells, which express FPR, and SHG-44 cells, which do not express FPR, in a xeno-graph model. We injected U-87 and SHG-44 cells subcutane-ously into the fl anks of athymic mice and measured the rate of tumor formation and growth. Tumor nodules appeared in all mice on day 4 after U-87 cell implantation and in all mice on day 10 after SHG-44 cell implantation. Tumors formed by U-87 cells grew more rapidly than did those formed by SHG-44 cells. By day 29 after implantation, all mice bearing U-87 tumors had become moribund. In contrast, all mice bearing SHG-44 tumors had become moribund by day 44 after implantation ( Fig. 5, B ). Thus, FPR-expressing glioblastoma cells appear to have a higher rate of tumorigenicity and growth in vivo in athymic mice than do glioma cells that do not express a func-tional FPR.

To determine the clinical relevance of this result, we exam-ined specimens derived from 33 surgically removed gliomas with various grades. FPR protein staining was detected in 11 of 14 grade III anaplastic astrocytoma specimens and six of six grade IV glioblastoma multiforme specimens. Microvessels and ne-crotic tumor cells were readily visible among FPR-positive intact tumor cells, as shown in a representative section from a grade IV glioblastoma multiforme tumor ( Fig. 5, C ). In contrast, only two of 13 less aggressive grade II astrocytoma specimens showed posi-tive FPR staining. Thus, FPR expression appears to be associated

Fig. 4. Formylpeptide receptor (FPR) activation, DNA synthesis, and levels of Bcl-2 and Bcl-xL in U-87 cells. A to C ) DNA synthesis. U-87 cells (8000 cells per well of 96-well plates) were pretreated with medium ( A ), 5 μ M cyclosporine H (CsH, an FPR inhibitor) ( B ), or 5 μ M Tyrphostin AG490 (AG490; a tyrosine kinase inhibitor that disrupts JAK/STAT signaling) ( C ) and then incubated in medium containing 0.5% fetal calf serum in the presence or absence of 1 μ M N -formyl-methionyl-leucyl-phenylalanine (fMLF) as indicated. Cells were then pulse-labeled with [ 3 H]thymidine (1 μ Ci per well) for the last 16 hours, and incorporated radioactivity was measured. *, P values range from .029 to <.001, compared with cells cultured in medium alone ( A ) or with cells stimulated with fMLF alone ( B and C ) (two-sided t test). D ) Expression of Bcl-2 and Bcl-xL in fMLF-treated U-87 cells. Cells were cultured in medium containing 0.5% fetal calf serum with fMLF as indicated, and then the levels of Bcl-2 and Bcl-xL proteins were measured by immunoblot analysis. In another experiment, U-87 cells were pretreated with PD98059, an inhibitor of the mitogen-activated protein kinase (MAPK) kinase 1 (MEK1) (50 μ M ) for 1 hour, and then the level of Bcl-2 was measured at 8 hours after stimulation with 1 μ M fMLF. β -Actin was used as a loading control.


Dow

nloaded from



with a majority of poorly differentiated primary human gliomas of grades III and IV.

FPR Knockdown by siRNA and Tumorigenicity of Glioblastoma Cells

To more precisely evaluate the role of FPR in glioblastoma tumorigenicity, we used siRNA to inhibit the expression and function of FPR in U-87 cells. After stable transfection of FPR siRNA into U-87 cells, the expression of FPR mRNA ( Fig. 6, A ) and in vitro fMLF-induced chemotaxis ( Fig. 6, B ) were almost completely abolished, compared with those of cells transfected with random siRNA (i.e., mock-transfected cells). In addition, the ability of fMLF to induce phosphorylation of ERK1/2 ( Fig. 6, C ) and of STAT3 ( Fig. 6, D ) was abrogated, the cell proliferation rate was slowed, and the cells did not respond to the growth-stimulating activity of fMLF, compared with mock-transfected cells ( Fig. 6, E ).

To examine whether FPR contributed to the tumorigenicity of U-87 cells in vivo, we injected U-87 cells that had been trans-fected with FPR siRNA into the fl anks of athymic mice. Tumor

nodules formed by U-87 cells transfected with FPR siRNA appeared later ( Fig. 7, A ), and the corresponding tumors grew more slowly than those formed by wild-type U-87 cells or by mock-transfected cells ( Fig. 7, B ). By day 42 after implantation, all mice implanted with wild-type or mock-transfected U-87 cells had died or had to be sacrifi ced because they carried large ne-crotic tumors. In contrast, all mice bearing tumors formed by FPR siRNA-transfected U-87 cells survived to at least day 72 after implantation ( Fig. 7, C ). These results indicate that deple-tion of FPR from U-87 cells markedly reduced their ability to form tumors in athymic mice and improved the survival rate of tumor-bearing mice. In addition, we transfected another human glioblastoma cell line, SNB75, which expresses functional FPR ( 7 ) , with FPR siRNA and determined the effect of FPR siRNA on the tumorigenicity of these cells. SNB75 cells trans-fected with FPR siRNA lost the ability to chemotactically respond to fMLF ( Fig. 7, D ). Tumors formed by SNB75 cells that were transfected with FPR siRNA and injected in nude mice grew more slowly than those formed by mock-transfected SNB75 cells. For example, 14 days after implantation, tumors formed by SNB75 cells transfected with FPR siRNA had a size of 12 mm 3

Fig. 5. Phenotype and tumorigenicity of glioma cells. A ) Expression of glial fi brillary acidic protein (GFAP; red), a glial cell differentiation marker, and vimentin (red), a marker of poorly differentiated astroglial cells, by U-87 and SHG-44 glioma cells. Glioma cells cultured on chamber slides were fi xed in ethanol, permeabilized with Triton X-100, incubated with normal goat serum to block potential nonspecifi c binding of antibodies to the cells, and then incubated with antibodies against GFAP and vimentin. The slides were then treated with an EnVision kit to visualize GFAP and vimentin under light microscopy (original amplifi cation = ×200). B ) Tumor formation and progression. U-87 or SHG-44 cells (5 × 10 6 cells in 100 μ L of phosphate-buffered saline [PBS]) were subcutaneously injected in the fl ank of each athymic mouse, and mice were examined for tumor formation on the indicated days. Each mouse received a single injection. Tumor size was expressed as the mean volume (in mm 3 ) of the tumors from fi ve mice. Error bars = 95% confi dence intervals. C ) Expression of the formylpeptide receptor (FPR) in primary tumors. Sections from 33 primary gliomas were stained with an antibody against the FPR (brown) and counterstained with hematoxylin (purple-blue). FPR protein staining was detected in 11 of 14 grade III anaplastic astrocytoma specimens, six of six grade IV glioblastoma multiforme specimens, and two of 13 grade II astrocytoma specimens. Representative sections of the FPR-positive specimens are shown, each with similar results. Arrows = FPR-positive cells in sections from a grade IV glioblastoma multiforme and a grade III anaplastic astrocytoma. In grade IV tumors: black arrowhead = a microvessel; white arrowhead = necrotic area in the tumor. Bar = 10 μ m.


Dow

nloaded from



(95% CI = 8 to 16 mm 3 ), and tumors formed by mock-transfected SNB75 cells had a size of 33 mm 3 (95% CI = 28 to 38 mm 3 ) ( P <.001). Thus, FPR appears to contribute to the growth of SNB75 glioblastoma in nude mice.

Production of Molecules with FPR Agonist Activity by Necrotic Glioblastoma Cells

Because a characteristic feature of malignant glioma is the presence of necrosis, even in relatively small lesions with vigor-ous neovascularization ( 26 ) , and because mitochondria of rup-tured cells contain chemotactic formylpeptides that appear to activate FPR in myeloid cells ( 27 ) , we investigated whether necrotic glioblastoma cells and tissues produced a natural agonist(s) recognized by FPR on glioblastoma cells. We found that U-87 cells and U-87 tumors formed in athymic mice released potent chemotactic activity for live U-87 cells ( Fig. 8, A and B ) and for ETFR cells, which overexpress FPR (data not shown). The FPR agonist activity released by necrotic U-87 cells and U-87 tumor tissues was blocked by an anti-FPR antibody or by

the FPR- specifi c antagonist tBoc-MLF ( 28 ) ( Fig. 8, B , and data not shown). Necrotic glioblastoma cell supernatant also induced a robust intracellular Ca 2+ mobilization in live U-87 cells ( Fig. 8, C ) and attenuated the ability of U-87 cells to respond to fMLF administered subsequently ( Fig. 8, D ). These results suggest that fMLF and the natural agonist activity contained in the superna-tants of necrotic U-87 cells may share the common GPCR, FPR, on U-87 ( 11 ) . We additionally observed that necrotic tumor su-pernatants inhibited the expression of FPR on cell surface of ETFR cells with an effi cacy comparable to that of bacterial fMLF at 10 3 n M ( Fig. 8, E ). Thus, necrotic glioblastoma cells appear to produce an FPR agonist activity that interacts with the FPR on live tumor cells.

D ISCUSSION

In this article, we show that FPR is selectively expressed by highly malignant glioma cells but not by less aggressive glioma cells or by normal human astrocytes. To our knowledge, this is the fi rst demonstration that activated FPR may contribute to the

Fig. 6. Short interfering RNA (siRNA) and the function of formylpeptide receptor (FPR) in U-87 cells. A ) Expression of FPR mRNA. The level of FPR mRNA was examined by reverse transcription – polymerase chain reaction (RT-PCR) in U-87 cells stably transfected with FPR siRNA constructs ( lane 3 ), wild-type (nontransfected) U-87 cells ( lane 1 ), and mock-transfected U-87 cells ( lane 2 ). RT-PCR product of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. B ) FPR-mediated chemotaxis. The ability of siRNA-transfected U-87 cells and mock-transfected U-87 cells to respond to N -formyl-methionyl-leucyl-phenylalanine (fMLF) by chemotaxis was examined by placing 5 × 10 4 tumor cells in the upper wells of the chemotaxis chamber and fMLF in the lower wells; the wells were separated by a 10- μ m (pore-size) polycarbonate fi lter coated with collagen type I. After incubation, cells that migrated across the fi lter were counted under light microscopy in three high-power fi elds (×400) for each experimental group. Data are the mean of three experiments. Error bars = 95% confi dence intervals. *, P values range from .001 to <.001, compared with mock-transfected cells (two-sided t test). C and D ) FPR-induced

phosphorylation of the extracellular signal-regulated kinases 1/2 (ERK1/2) and the signal transducers and activators of transcription 3 (STAT3). Mock-transfected U-87 cells and siRNA transfected U-87 cells were stimulated with 10 2 n M fMLF, as indicated, and phosphorylation of ERK1/2 ( C ) and STAT3 ( D ) was examined by immunoblotting. The protein bands are labeled as follows: p-ERK = phosphorylated ERK1/2; ERK = total ERK1/2; P-Tyr705 and p-Ser727 = phosphorylated Tyr-705 and Ser-727 residues in STAT3, respectively; STAT3 = total STAT3. E ) DNA synthesis. Mock-transfected U-87 cells and siRNA-transfected U-87 cells (8000 cells per well in a 96-well plate) were cultured in medium containing 0.5% fetal calf serum at 37 °C in the presence or absence of 1 μ M fMLF. Cells were then pulse-labeled with [ 3 H]thymidine (1 μ Ci per well) for the last 16 hours (n = 4), and the incorporated radioactivity was measured as counts per minute (cpm). The amount of DNA synthesized refl ects the level of cell proliferation. *, P values range from .023 to .004 compared with cells treated with medium alone. Data are the mean of three experiments. Error bars = 95% confi dence intervals.


Dow

nloaded from



progression of highly malignant gliomas by mediating tumor cell chemotaxis, proliferation, and production of VEGF in response to an agonist(s) potentially produced by necrotic tumor cells.

Because FPR was originally detected in cells of the immune system and interacts with bacterial chemotactic peptides, this receptor was hypothesized to participate in a host defense mecha-nism against microbial infection ( 1 ) . This hypothesis was sup-ported by reduced antibacterial responses in mice depleted of the counterpart murine receptor FPR1 ( 29 ) . However, FPR has been more recently reported to also interact with host- derived chemo-tactic peptides, including formylpeptides potentially released by mitochondria, annexin I produced by activated epithelia, and a neutrophil granule protein, cathepsin G ( 1 , 9 , 27 , 30 , 31 ) . In addi-tion, functional FPR has been detected in cells of nonhematopoi-etic origin, such as lung epithelial cells and hepatocytes ( 30 , 32 ) . These fi ndings have indicated that FPR may be involved in a broader spectrum of pathophysiologic processes that also include infl ammation and immunity. Our present study further extended the functional scope of FPR to its potential role in promoting the growth of malignant human glioma.

A hallmark in the progression of malignant tumors is increased angiogenesis, which has been attributed to the aberrant produc-tion of angiogenic factors. One of the most potent angiogenic factors produced in solid tumors is VEGF, which not only in-duces endothelial cell migration, proliferation, and tubule forma-tion but also increases microvascular permeability, which may facilitate dissemination of malignant tumor cells ( 33 – 35 ) . Anti-angiogenic intervention with VEGF antibodies or by VEGF withdrawal results in endothelial cell apoptosis and inhibition of tumor growth ( 36 , 37 ) . Malignant gliomas, notably glioblastoma multiforme, are characterized by a high degree of vascularity and the production of copious amounts of VEGF. Hypoxia may further increase the production of VEGF in tumor cells that

surround regions of necrosis, even in the early stages of tumor progression ( 38 , 39 ) . Hypoxia promotes VEGF gene transcription through nuclear translocation of HIF-1 α . HIF-1 α protein is overexpressed and undergoes enhanced nuclear translocation in a variety of human malignant tumors, and increased levels of HIF-1 α are associated with vigorous vascularization and tumor progression ( 17 , 40 ) . Although hypoxia induces the nuclear trans-location of HIF-1 α in many cell types, growth factors and genetic abnormalities frequently detected in human cancer can also in-crease the level of HIF-1 α protein, its DNA binding activity, and the expression of VEGF ( 41 , 42 ) . In this study, we have demon-strated, to our knowledge for the fi rst time, that activation of FPR in glioblastoma cells can promote the nuclear translocation of HIF-1 α and also increase the expression of VEGF mRNA and protein.

We found that MAPKs, including ERK1/2, p38, and JNK, were phosphorylated in FPR-expressing glioblastoma cells that were activated by peptide agonists, consistent with previously re-ported results obtained with myeloid cells ( 43 ) . ERK1/2 MAPK may play an important role in promoting endothelial cell prolif-eration, the expression of VEGF, and the resultant angiogenic process ( 44 ) . We showed that inhibition of FPR-mediated ERK1/2, but not p38, activity in U-87 cells reduced the levels of VEGF mRNA induced by FPR agonists and that inhibition of MEK1, but not p38, activity completely blocked FPR agonist – triggered nuclear translocation of HIF-1 α . Thus, the ERK1/2 pathway appears to be crucial for FPR agonist – induced VEGF expression in U-87 cells. However, it has also been reported ( 45 ) that a p38 inhibitor blocks prostaglandin E1 – stimulated VEGF synthesis in osteoblast-like cells, suggesting that in different cell types ERK1/2 and p38 may be differentially associated with the signaling cascade that promotes transcription of the VEGF gene. Further research to more clearly defi ne the identity of signaling molecules

Wild type

Mock

siRNA

Days after implantation

Tum

or s

ize

(mm

3 )

20

40

60

80

100

30201000

Wild type

Mock

siRNA

Days post-transplantation

% tu

mor

form

atio

n

A B

D

0

200

400

600

800

1000

1200

Day 30 Day 38

0

40

80

120

160

No.

of m

igra

ted

cells

Mock

siRNA

fMLF (nM)

*

*

10-8 10-7 10-60 25 50 750

25

50

75

100

Wild type

Mock

siRNA

Days after implantation

% s

urvi

val

C

M

Fig. 7. Formylpeptide receptor (FPR) short interfering RNA (siRNA) and glioblastoma progression in athymic mice. A ) Tumor formation. Nontransfected (wild-type), mock-transfected, or FPR siRNA-transfected U-87 cells (at 1 × 10 6 cells in 100 μ L of phosphate-buffered saline [PBS] per mouse) were injected subcutaneously into the fl anks of athymic mice (10 mice per group). Mice were examined for tumor formation as indicated. B ) Tumor growth. Tumor size 30 and 38 days after implantation of U-87 cells is presented as the mean volume (mm 3 ) of tumors from 10 mice per group, as described for panel A . Error bars = 95% confi dence intervals. C ) Survival of tumor-bearing mice. Mice are as described for panel A , with 10 mice per group. The numbers of mice at risk as calculated by Kaplan – Meier plot are as follows: on day 26, wild type = 8 (survival rate = 80%, 95% CI = 40% to 100%), mock = 6 (survival rate = 60%, 95% CI = 30% to 90%), and siRNA = 10 (survival rate = 100%); on day 32, wild type = 2 (survival rate = 20%, 95% CI = 0% to 40%, mock = 0 (survival rate = 0), and siRNA = 10 (survival rate = 100%); on day 72, wild type = 0 (survival rate = 0), mock = 0 (survival rate = 0), and siRNA = 10 (survival rate 100%). D ) N -Formyl-methionyl-leucyl-phenylalanine (fMLF) – induced chemotaxis in SNB75 cells stably transfected with FPR siRNA or mock-transfected. Chemotaxis induced by 10 − 8 , 10 − 7 , or 10 − 6 M fMLF was assessed by seeding the cells into the upper wells of the chemotaxis chamber and fMLF in the lower wells; wells were separated by a 10- μ m (pore size) polycarbonate fi lter. Medium (M) was used as the control. After incubation at 37 °C for 4.5 hours, the number of cells in three high-power fi elds (×400) was counted, and data are expressed as the mean. Error bars = 95% confi dence intervals. *, P values range from .006 to <.001 (two-sided t test).


Dow

nloaded from



Fig. 8. Formylpeptide receptor (FPR) agonist activity in necrotic tumor supernatants. A ) Chemotactic activity. Supernatants of U-87 cells after three cycles of freezing and thawing (Necrotic) were assayed for chemotactic activity with live U-87 cells. Supernatants from apoptotic U-87 cells (generated by treatment for 6 hours with 0.5 μ M staurosporine) or from cells under normal culture conditions (Live) were used as controls. The number of cells in three high-power fi elds (×400) was counted, and data are expressed as the mean. Error bars = 95% confi dence intervals. * indicates statistically signifi cantly increased cell migration, compared with control medium ( P values range from .002 to <.001). Bar 1 = medium; bar 2 = 10 2 n M N -formyl-methionyl-leucyl-phenylalanine (fMLF); bars 3 = tumor-cell supernatant containing total protein at 1.7 mg/mL; bars 4 = tumor-cell supernatant containing total protein at 5 mg/mL; bars 5 = tumor-cell supernatant containing total protein at 15 mg/mL. B ) Chemotactic activity of supernatants of tumor tissue extracts. U-87 tumor tissues from mice were minced and subjected to repeated cycles of freezing and thawing, and the supernatants were assessed for chemotactic activity on live U-87 cells, which were incubated in the presence ( open bars ) or absence ( solid bars ) of t- butyloxycarbonyl-methionyl-leucyl-phenylalanine at 37 °C for 30 minutes.

Bars 1 = no dilution of supernatant; bars 2 = 1 : 10 3 dilution; bars 3 = 1 : 10 2 dilution; bars 4 = 1 : 50 dilution. *, Bars 2 , P = .003; bars 3 and 4 , P values range from .001 to <.001, compared with untreated cells. C ) Ca 2+ fl ux induced in U-87 cells by diluted necrotic tumor cell supernatants or fMLF, as indicated. U-87 cells were loaded with Fura-2 acetoxymethyl ester at room temperature and then were transferred into quartz cuvettes, and cuvettes were placed in a fl uorescence spectrometer. Stimulants (i.e., fMLF or tumor supernatants) were added to each cuvette, and the intensity of the fl uorescence was measured by use of the ratio of the absorbance at 340 nm to the absorbance at 380 nm and calculated with an FL WinLab program. D ) Desensitization of Ca 2+ fl ux in U-87 cells. U-87 cells loaded with Fura-2 acetoxymethyl ester were placed in a cuvette, and then cells were stimulated with diluted necrotic tumor cell supernatant, followed 100 seconds later by fMLF. The intensity of the fl uorescence was measured by use of the ratio of the absorbance at 340 nm to the absorbance at 380 nm. E ) Regulation of FPR expression (as a percentage of FPR-positive cells) on ETFR cells by necrotic tumor cell supernatants. The level of FPR on the surface of ETFR cells, treated as indicated, was measured by fl ow cytometry with an anti-FPR antibody ( α -FPR) or normal immunoglobulin G, as a control. FLI = fl uorescence intensity.

associated with activated FPR in glioma cells that may promote the production of angiogenic factors will be important in the design of antiangiogenic therapy for malignant human gliomas.

We found that FPR may increase the survival of glioblastoma cells under suboptimal culture conditions by increasing the levels of Bcl-2. In a variety of cell types, the effect of growth factors on survival is dependent on the relative levels of pro- versus anti-apoptotic members of the Bcl-2 family. Bcl-2, a widely studied antiapoptotic protein, is thought to interfere with the release of cytochrome c from mitochondria and the activation of procas-pase 9 ( 46 , 47 ) . In a murine B-cell lymphoma cell line ( 48 ) , activation of the MEK/ERK pathway is associated with Bcl-2 expression. This result is in accordance with our fi ndings that a MEK/ERK inhibitor reduced the level of fMLF-induced Bcl-2 in human U-87 cells. Thus, we identifi ed an additional role of FPR, i.e., mediation of tumor cell survival through the MEK/ERK- dependent signaling cascade. Our results suggest that FPR may also be considered a target for the design of agents that inhibit glioblastoma cell survival and proliferation. In fact, the ability of the MEK1 inhibitor PD98059 to inhibit the growth and clono-genicity of acute myeloid leukemia cells has been attributed to its reduction of the level of Bcl-2 in these cells ( 49 , 50 ) .

We also demonstrated that FPR agonists, such as fMLF, activate the transcriptional factor STAT3 in U-87 cells. STAT3 is a key signaling molecule downstream of receptors for many cytokines and growth factors ( 51 ) . In malignant tumor cells, STAT3 is in an activated state and plays a critical role in onco-genesis, the production of angiogenic factors, and cell survival ( 52 – 54 ) . Phosphorylation of STAT3 at Tyr-705 induces the dimerization of STAT3, the nuclear translocation of the STAT3 dimer, and its binding to the promoter regions of target genes ( 55 ) . Phosphorylation of STAT3 at Ser-727 in the carboxyl- terminal region maximizes the transcriptional activity of STAT3 ( 20 , 56 ) . Although disruption of the STAT3 pathway does not cause death of normal cells in in vitro ( 57 ) and in animal ( 22 ) models, activated STAT3 is essential for tumor cell survival ( 54 ) . Although we did not detect high levels of consti-tutively active STAT3 in U-87 cells, the FPR agonist fMLF i ncreased the level of phosphorylation at both Tyr-705 and Ser-727 residues. In addition, because the JAK/STAT inhibitor Tyrphostin AG490 inhibited the growth- stimulating activity of fMLF, the JAK/STAT pathway appears to be associated with activated FPR, as also reported for selected chemokine GPCRs ( 18 , 19 ) .


Dow

nloaded from



Malignant tumors exploit their microenvironment to favor their survival, growth, invasion, and metastasis ( 58 ) . For instance, tumor cells often produce aberrant levels of growth factors that stimulate cell surface receptors to increase cell proliferation in an autocrine manner. Tumor cells also produce high levels of VEGF constitutively or in response to stimulation that recruits endothe-lial cells and promotes endothelial cell proliferation and vascu-larization. In addition, malignant tumor cells express receptors that interact with agonists that are present in the vicinity of the tumor or produced by distant organs to increase tumor cell motil-ity and thus to favor tumor cell invasion and metastasis. Chemo-kine receptors, such as CXCR4 and CCR7, have been implicated in promoting tumor metastasis, presumably by increasing the chemotaxis and extravasation of tumor cells in response to lo-cally produced chemokine ligands ( 2 , 4 , 59 ) . The chemokine re-ceptor CXCR4, expressed by a majority of glioma cell lines, mediates tumor cell migration and supports cell survival, pre-sumably in response to the ligand SDF-1 α , which is present in tumor tissues ( 60 ) . However, because CXCR4 is also expressed in normal astrocytes, it may not be a good biomarker for differ-entiating normal astrocytes from malignant astrocytes or for dis-tinguishing less aggressive tumor cells from highly aggressive malignant tumor cells. In contrast, FPR is not widely expressed in glioma cell lines or in normal glial cells but, rather, is expressed in more highly malignant glioblastoma cells and contributes to their tumorigenicity in vivo. FPR is also detected in a majority of primary grade III anaplastic astrocytoma and grade IV glioblas-toma multiforme specimens that we examined. Identifi cation of FPR agonist activity in the supernatants of necrotic tumor cells provides evidence that this receptor may interact with host- derived agonists produced in tumor lesions, presumably in the necrotic area frequently associated with highly malignant gliomas or in surrounding tissues that are compressed by growing tumor in a limited anatomical compartment. It is thus plausible that FPR in live tumor cells may serve as a sensor for the agonists pro-duced in a “ paracrine ” manner in the tumor microenvironment to promote cell migration, to support cell survival and proliferation, to activate transcription of VEGF mRNA, and to increase the production of VEGF.

Further study is required to more precisely defi ne the relation-ship between the FPR expression and the progression of human primary gliomas and to identify the mechanistic basis for the control of FPR expression in highly malignant human glioma cells. In addition, the pathogenesis of human gliomas is likely to be complex, and FPR may not be the sole factor that regulates the progression of malignant gliomas. Indeed, our observation that there were three FPR-protein-negative tumors of the 14 primary grade III anaplastic astrocytoma specimens examined suggests that factors other than FPR also participate in the development of malignant human gliomas. Also, the relationship between FPR expression and the survival of glioma patients after treatment remains to be established. Nevertheless, the present study impli-cates the role of FPR in the rapid progression of highly malignant human gliomas and thus raises the possibility that FPR may be a candidate molecular target for developing novel therapeutics to treat gliomas.

REFERENCES

(1) Le Y, Oppenheim JJ, Wang JM. Pleiotropic roles of formyl peptide recep-tors. Cytokine Growth Factor Rev 2001 ; 12 : 91 – 105.

(2) Wang JM, Deng X, Gong W, Su S. Chemokines and their role in tumor growth and metastasis. J Immunol Methods 1998 ; 220 : 1 – 17.

(3) Muller A, Homey B, Soto H, Ge N, Catron D, Buchanan ME, et al. Involvement of chemokine receptors in breast cancer metastasis. Nature 2001 ; 410 : 50 – 6.

(4) Wang JM, Chertov O, Proost P, Li JJ, Menton P, Xu L, et al. Purifi cation and identifi cation of chemokines potentially involved in kidney-specifi c metastasis by a murine lymphoma variant: induction of migration and NF κ B activation. Int J Cancer 1998 ; 75 : 900 – 7.

(5) Oh JW, Drabik K, Kutsch O, Choi C, Tousson A, Benveniste EN. CXC chemokine receptor 4 expression and function in human astroglioma cells. J Immunol 2001 ; 166 : 2695 – 704.

(6) Dorf ME, Bermana MA, Tanabea S, Heesena M, Luoa Y. Astrocytes express functional chemokine receptors. J Neuroimmunol 2000 ; 111 : 109 – 21.

(7) Le Y, Hu J, Gong W, Shen W, Li. B, Dunlop NM, et al. Expression of func-tional formyl peptide receptors by human astrocytoma cell lines. J Neuroim-munol 2000 ; 111 : 102 – 8.

(8) Prossnitz ER, Ye RD. The N-formyl peptide receptor: a model for the study of chemoattractant receptor structure and function. Pharmacol Ther 1997 ; 74 : 73 – 102.

(9) Murphy PM. The molecular biology of leukocyte chemoattractant receptors. Annu Rev Immunol 1994 ; 12 : 593 – 633.

(10) Salcedo R, Wasserman K, Young HA, Grimm MC, Howard OM, Anver MR, et al. Vascular endothelial growth factor and basic fi broblast growth factor induce expression of CXCR4 on human endothelial cells: in vivo neovascularization induced by stromal-derived factor-1 α . Am J Pathol 1999 ; 154 : 1125 – 35.

(11) Le Y, Gong W, Li B, Dunlop NM, Shen W, Su SB, et al. Utilization of two seven-transmembrane, G protein-coupled receptors, formyl peptide receptor-like 1 and formyl peptide receptor, by the synthetic hexapeptide WKYMVm for human phagocyte activation. J Immunol 1999 ; 163 : 6777 – 84.

(12) Zheng MH, Xu J, Robbins P, Pavlos N, Wysocki S, Kumta SM, et al. Gene expression of vascular endothelial growth factor in giant cell tumors of bone. Hum Pathol 2000 ; 31 : 804 – 12.

(13) Le Y, Iribarren P, Zhou Y, Gong W, Hu J, Zhang X, et al. Silencing the formylpeptide receptor FPR by short-interfering RNA. Mol Pharmacol 2004 ; 66 : 1022 – 8.

(14) Ponten J, Macintyre EH. Long term culture of normal and neoplastic human glia. Acta Pathol Microbiol Scand 1968 ; 74 : 465 – 86.

(15) Vivanco I, Sawyers CL. The 3-kinase-AKT pathway in human cancer. Nat Rev Cancer 2002 ; 2 : 489 – 501.

(16) Browning DD, Pan ZK, Prossnitz ER, Ye RD. Cell type- and developmental stage-specifi c activation of NF- κ B by fMet-Leu-Phe in myeloid cells. J Biol Chem 1997 ; 272 : 7995 – 8001.

(17) Harris AL. Hypoxia — a key regulatory factor in tumor growth. Nat Rev Cancer 2002 ; 2 : 38 – 47.

(18) Wong M, Fish EN. RANTES and MIP-1 α activate stats in T cells. J Biol Chem 1998 ; 273 : 309 – 14.

(19) Mellado M, Rodriguez-Frade JM, Manes S, Martinez AC. Chemokine sig-naling and functional responses: the role of receptor dimerization and TK pathway activation. Annu Rev Immunol 2001 ; 19 : 397 – 421.

(20) Wen Z, Zhong Z, Darnell JE Jr. Maximal activation of transcription by Stat1 and Stat3 requires both tyrosine and serine phosphorylation. Cell 1995 ; 82 : 241 – 50.

(21) Wen Z, Darnell JE Jr. Mapping of Stat3 serine phosphorylation to a single residue (727) and evidence that serine phosphorylation has no in-fl uence on DNA binding of Stat1 and Stat3. Nucleic Acids Res 1997 ; 25 : 2062 – 7.

(22) Meydan N, Grunberger T, Dadi H, Shahar M, Arpaia E, Lapidot Z, et al. Inhibition of acute lymphoblastic leukaemia by a Jak-2 inhibitor. Nature 1996 ; 379 : 645 – 8.

(23) Nielsen M, Kaltoft K, Nordahl M, Ropke C, Geisler C, Mustelin T, et al. Constitutive activation of a slowly migrating isoforms of Stat3 in myco-sis fungoides: tyrphostin AG490 inhibits Stat3 activation and growth of mycosis fungoides tumor cell lines. Proc Natl Acad Sci U S A 1997 ; 94 : 6764 – 9.

(24) Dahlstrand J, Lardelli M, Lendahl U. Nestin mRNA expression correlates with the central nervous system progenitor cell state in many, but not all, regions of developing central nervous system. Dev Brain Res 1995 ; 84 : 109 – 29.


Dow

nloaded from



(25) Rutka JT, Murakami M, Dirks PB, Hubbard SL, Becker LE, Fukuyama K, et al. Role of glial fi laments in cells and tumors of glial origin: a review. J Neurosurg 1997 ; 87 : 420 – 30.

(26) Reijneveld JC, Voest EE, Taphoorn MJ. Angiogenesis in malignant primary and metastatic brain tumors. J Neurol 2000 ; 247 : 597 – 608.

(27) Carp H. Mitochondrial N-formylmethionyl proteins as chemoattractants for neutrophils. J Exp Med 1982 ; 155 : 264 – 75.

(28) Le Y, Sun R, Ying G, Iribarren P, Wang JM. The role of peptide receptors in microbial infection and infl ammation. Curr Med Chem Anti-Infect Agents 2003 ; 2 : 83 – 93.

(29) Gao JL, Lee EJ, Murphy PM. Impaired antibacterial host defense in mice lacking the N-formylpeptide receptor. J Exp Med 1999 ; 189 : 657 – 62.

(30) Rescher U, Danielczyk A, Markoff A, Gerke V. Functional activation of the formyl peptide receptor by a new endogenous ligand in human lung A549 cells. J Immunol 2002 ; 169 : 1500 – 4.

(31) Sun R, Iribarren P, Zhang N, Zhou Y, Gong W, Cho EH, et al. Identifi ca-tion of neutrophil granule protein Cathepsin G as a novel chemotactic ag-onist for the G protein coupled formyl peptide receptor FPR. J Immunol 2004 ; 173 : 428 – 36.

(32) McCoy R, Haviland DL, Molmenti EP, Ziambaras T, Wetsel RA, Perlmutter DH. N-formylpeptide and complement C5a receptors are expressed in liver cells and mediate hepatic acute phase gene regulation. J Exp Med 1995 ; 182 : 207 – 17.

(33) Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995 ; 1 : 27 – 31.

(34) Ferrara N. Role of vascular endothelial growth factor in the regulation of angiogenesis. Kidney Int 1999 ; 56 : 794 – 814.

(35) Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature 2000 ; 407 : 249 – 57.

(36) Borgstrom P, Bourdon MA, Hillan KJ, Sriramarao P, Ferrara N. Neutraliz-ing anti-vascular endothelial growth factor antibody completely inhibits an-giogenesis and growth of human prostate carcinoma micro tumors in vivo. Prostate 1998 ; 35 : 1 – 10.

(37) Benjamin LE, Keshet E. Conditional switching of vascular endothelial growth factor (VEGF) expression in tumors: induction of endothelial cell shedding and regression of hemangioblastoma-like vessels by VEGF with-drawal. Proc Natl Acad Sci U S A 1997 ; 94 : 8761 – 6.

(38) Chan AS, Leung SY, Wong MP, Yuen ST, Cheung N, Fan YW, et al. Expres-sion of vascular endothelial growth factor and its receptors in the anaplastic progression of astrocytoma, oligodendroglioma, and ependymoma. Am J Surg Pathol 1998 ; 22 : 816 – 26.

(39) Prodos MD, Levin V. Biology and treatment of malignant glioma. Semin Oncol 2000 ; 27 : 1 – 10.

(40) Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DZ, Zagzag D, et al. Overexpression of hypoxia-inducible factor 1 α in common human cancers and their metastases. Cancer Res 1999 ; 59 : 5830 – 5.

(41) Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, et al. The tumor suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999 ; 399 : 271 – 5.

(42) Zundel W, Schindler C, Haas-Kogan D, Koong A, Kaper F, Chen E, et al. Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 2000 ; 14 : 391 – 6.

(43) Rane MJ, Carrithers S, LArthur JM, Klein JB, McLeish KR. Formyl peptide receptors are coupled to multiple mitogen-activated protein kinase cascades by distinct signal transduction pathways: role in activation of reduced nico-tinamide adenine dinucleotide oxidase. J Immunol 1997 ; 159 : 5070 – 8.

(44) Pages G, Milanini J, Richard DE, Berra E, Gothie E, Vinals F, et al. Sig-naling angiogenesis via p42/p44 MAP kinase cascade. Ann N Y Acad Sci 2000 ; 902 : 187 – 200.

(45) Tokuda H, Kozawa O, Miwa M, Uematsu T. p38 mitogen-activated protein (MAP) kinase but not p44/p42 MAP kinase is involved in prosta-glandin E 1 -induced vascular endothelial growth factor synthesis in osteoblasts. J Endocrinol 2001 ; 170 : 629 – 38.

(46) Gross A, McDonnell JM, Dorsmeyer SJ. BCL-2 family members and the mitochondria in apoptosis. Genes Dev 1999 ; 13 : 1899 – 911.

(47) Murphy KM, Ranganathan V, Farnsworth ML, Kavallaris M, Lock RB. Bcl-2 inhibits Bax translocation from cytosol to mitochondria during drug-induced apoptosis of human tumor cells. Cell Death Differ 2000 ; 7 : 102 – 11.

(48) Kurland JF, Voehringer DW, Meyn RE. The MEK/ERK pathway acts up-stream of NF κ B1 (p50) homodimers activity and Bcl-2 expression in a mu-rine B-cell lymphoma cell line. MEK inhibition restores radiation-induced apoptosis. J Biol Chem 2003 ; 278 : 32465 – 70.

(49) Milella M, Estrov Z, Kornblau SM, Carter BZ, Konopleva M, Tari A, et al. Synergistic induction of apoptosis by simultaneous disruption of the Bcl-2 and MEK/MAPK pathways in acute myelogenous leukemia. Blood 2002 ; 99 : 3461 – 4.

(50) Milella M, Kornblau SM, Estrov Z, Carter BZ, Lapillonne H, Harris D, et al. Therapeutic targeting of the MEK/MAPK signal transduction module in acute myeloid leukemia. J Clin Invest 2001 ; 108 : 851 – 9.

(51) Heim HH. The Jak-STAT pathway: cytokine signalling from the receptor to the nucleus. J Recept Signal Transduct Res 1999 ; 19 : 75 – 120.

(52) Garcia R, Jove R. Activation of STAT transcription factors in oncogenic tyrosine kinase signaling. J Biomed Sci 1998 ; 5 : 79 – 85.

(53) Bromberg JF, Wrzeszczynska MH, Devgan G, Zhao Y, Pestell RG, Albancese C, et al. Stat3 as an oncogene. Cell 1999 ; 98 : 295 – 303.

(54) Catlett-Falcone R, Landowski TH, Oshiro MM, Turkson J, Levitzki A, Savino R, et al. Constitutive activation of Stat3 signaling confers resistance to apoptosis in human U266 myeloma cells. Immunity 1999 ; 10 : 105 – 15.

(55) Bowman T, Garcia R, Turkson J, Jove R. STATs in oncogenesis. Oncogene 2000 ; 19 : 2474 – 88.

(56) Yokogami K, Wakisaka S, Avruch J, Reeves SA. Serine phosphorylation and maximal activation of STAT3 during CNTF signaling is mediated by the rapamycin target mTOR. Curr Biol 2000 ; 10 : 47 – 50.

(57) Turkson J, Bowman T, Garcia R, Caldenhoven E, De Groot RP, Jove R. Stat3 activation by Src induces specifi c gene regulation and is required for cell transformation. Mol Cell Biol 1998 ; 18 : 2545 – 52.

(58) Liotta LA, Kohn EC. The microenvironment of the tumor-host interface. Nature 2001 ; 411 : 375 – 9.

(59) Murphy PM. Chemokines and the molecular basis of cancer metastasis. N Engl J Med 2001 ; 345 : 833 – 5.

(60) Zhou Y, Larsen PH, Hao C, Yong VW. CXCR4 is a major chemokine receptor on glioma cells and mediates their survival. J Biol Chem 2002 ; 51 : 49481 – 7.

NOTES

Ye Zhou and Xiuwu Bian contributed equally to the study. The content of this publication does not necessarily refl ect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial prod-ucts, or organizations imply endorsement by the U.S. government. The publisher or recipient acknowledges the right of the U.S. government to retain a nonexclu-sive, royalty-free license in and to any copyright covering the article.

This project was funded in part with federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. NO1-CO-12400. We thank Dr. J. J. Oppenheim for critical review of this manuscript.

Manuscript received August 31, 2004; revised March 18, 2005; accepted April 20, 2005.


Dow

nloaded from


Formylpeptide Receptor FPR and the Rapid Growth of Malignant Human Gliomas

Documents

Transcript of Formylpeptide Receptor FPR and the Rapid Growth of Malignant Human Gliomas