Purpose: The role of CXCR4 in bone marrow localization of neuroblastoma cells has been recently proposed. The aim of this study was to investigate the expression and chemotactic functionality of CXCR4 in human metastatic neuroblastoma cells isolated from the bone marrow and, for comparison, in a panel of neuroblastoma cell lines.

Experimental Design: CXCR4 expression and chemotactic functionality were investigated in metastatic neuroblastoma cells isolated from patient bone marrow and in neuroblastoma cell lines. The former cells were isolated as CD45 or GD2+ cells by immunomagnetic bead manipulation. Chemotactic assays were done in a transwell system. Regulator of G protein signaling expression was investigated by reverse transcription-PCR.

Results: Metastatic neuroblastoma cells consistently expressed CXCR4, which was also detected in 5 of 10 neuroblastoma cell lines. CXCL12 did not stimulate the chemotaxis of primary tumor cells or cell lines in either normoxia or hypoxia, irrespective of CXCR4 up-regulation detected under the latter condition. Accordingly, neuroblastoma cells failed to modulate filamentous actin and to activate mitogen-activated protein kinase upon treatment with CXCL12. RGS16 mRNA was consistently expressed in primary tumor cells and cell lines, but its down-regulation by RNA interference did not restore CXCR4 chemotactic functionality.

Conclusions: These results show unambiguously that CXCR4 expressed in human metastatic neuroblastoma cells is not functional and do not support the clinical use of CXCR4 antagonists to prevent neuroblastoma metastasis.

Primary tumor cells of different origin or cell lines thereof express CXCR4, migrate in vitro to its ligand CXCL12 and metastasize in vivo to anatomic sites where CXCL12 is expressed (110). Accordingly, CXCR4 up-regulation has been shown in some human metastatic versus primary tumors (1113).

CXCR4 belongs to the wide family of G protein-coupled receptors. Regulator of G protein signaling (RGS) proteins terminate G protein-coupled receptor–mediated signaling by shortening the lifetime of Gα-GTP (14) and dampening mitogen-activated protein kinase (MAPK) phosphorylation (1518). RGS1, RGS3, RGS13, RGS16, and RGS18 have been shown to turn off CXCR4 signaling in different experimental systems (1921).

Preclinical studies indicate that metastatic spreading of CXCR4+ tumor cells can be inhibited by anti-CXCR4 antibodies or CXCR4 antagonists (1, 3, 4, 22). These results delineate a novel strategy for prevention and/or treatment of metastases that will certainly have an effect on the clinical practice of oncology.

Neuroblastoma is a pediatric tumor of neuroectodermal origin whose overall 5 year survival is 45% according to a recent European study (23). Therefore, novel therapeutic approaches for neuroblastomas with poor prognoses are needed. Among them, inhibition of dissemination of tumor cells from primary to metastatic sites may represent an attractive strategy.

The bone marrow is the major site of neuroblastoma metastasis, both at diagnosis and at relapse. A previous study done using neuroblastoma cell lines and a recent report on primary neuroblastoma tumors supported the potential involvement of CXCR4 in bone marrow localization of metastatic neuroblastoma cells (5, 9).

Here, we have investigated CXCR4 expression and function in primary metastatic neuroblasts isolated from the bone marrow of patients with neuroblastoma in comparison with a panel of neuroblastoma cell lines.

Patient samples. This investigation was done following approval by a local institutional review board. The criteria for diagnosis and evaluation of disease extension have been reported elsewhere (24). An aliquot of bone marrow aspirates done for diagnostic purposes was obtained from six patients with stage 4 disease at diagnosis. The median age was 3.3 years (range, 8 months to 5.9 years) and all of the patients had consistent bone marrow infiltration (2+/3+), as assessed by morphologic analysis. At present (8-10 months after diagnosis) one of the patients has died of disease and four, including an infant, have relapsed.

Cell separation and culture. Bone marrow aspirates were depleted of erythrocytes by osmotic lysis and washed in PBS. CD45− neuroblasts were isolated from bone marrow cell suspensions by incubation first with CD45 (Caltag, Burlingame, CA) monoclonal antibody (mAb) and subsequently with immunomagnetic beads coated with antimouse immunoglobulin antibodies, according to the instructions of the manufacturer (Miltenyi Biotec, Bergisch Gladbach, Germany). In some experiments, neuroblasts were positively selected as GD2+ cells using the same technique.

ACN, GI-ME-N, GI-LI-N, GI-CA-N, LAN-1, LAN-5, HTLA-230, SH-SY-5Y, IMR-32, SK-N-SH, and SK-N-BE-2c neuroblastoma cell lines, and the BRG Burkitt's lymphoma and Tanque pre–B leukemia cell lines, were cultured in RPMI 1640 (Sigma, St. Louis, MO) supplemented with l-glutamine, penicillin/streptomycin, nonessential amino acids, and 10% FCS (Sigma). In some experiments, neuroblastoma cell lines were cultured overnight in an anaerobic work station incubator, as described (25), to achieve 1% oxygen tension.

Monoclonal antibodies and flow cytometry. Surface staining was done as previously described (26). The source of anti-GD2 mAb (IgG2a) was the supernatant of the ME361-S2a murine hybridoma, purchased from American Type Culture Collection (Manassas, VA). Anti-CXCR4, CD45, FITC-, or phycoerythrin-conjugated goat anti-mouse IgG subclass, as well as isotype-matched control mAbs were purchased from Caltag. Anti-RGS16, that works exclusively in immunofluorescence, was purchased from Santa Cruz Biotechnology, Inc., Santa Cruz, CA.

Cells were scored using a FACScan analyzer (Becton Dickinson, San Jose, CA) and data were processed using CellQuest software (Becton Dickinson). The threshold line was based on the maximum staining obtained with irrelevant isotype-matched mAb, used at the same concentration as test mAb. Negative cells were defined such that <1% of cells stained positive with control mAbs. Cells labeled with test antibody that were brighter than those stained with isotypic control antibody were defined as positive.

Immunocytochemistry. Cells (2 × 105/slide) were cytocentrifuged, air-dried, and fixed in 10% buffered formalin for 10 minutes. Endo genous peroxidase activity was blocked after 10 minutes of incubation at room temperature with methanol containing 3% H2O2. Cytospins were then incubated overnight at 4°C with optimal amounts of anti-CXCR4 mAb or of isotype- and subclass-matched mouse immunoglobulin. Cytospins were subsequently washed twice in Optimax Wash Buffer (Menarini Diagnostics, Firenze, Italy) and incubated for 30 minutes at room temperature with Dako Envision system horseradish peroxidase mouse (Dako, Glostrup, Denmark). After washing in Optimax wash buffer, peroxidase activity was detected by incubating cytospins for 6 to 10 minutes at room temperature with Dako liquid diaminobenzidine substrate chromogen system (Dako). Cytospins were finally counterstained with Mayer's hematoxylin (Sigma).

Immunofluorescence. RGS-16 expression in neuroblastoma cell lines was investigated by immunofluorescence. Cells (50,000/slide) were cytocentrifuged, dried in air, and fixed in ice-acetone for 10 minutes. Cytospins were blocked at room temperature with PBS containing 10% rabbit serum for 1 hour and subsequently incubated with an optimal amount of anti-RGS16 goat polyclonal IgG (Santa Cruz Biotechnology). After washing with PBS, the slides were incubated with rabbit anti-goat IgG conjugated with Alexa Fluor-488 (Molecular Probes, Eugene, OR) for 20 minutes at 4°C. The cytospins were then washed with PBS, air-dried, mounted with Vectashield mounting medium with 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA) and observed under a Nikon epifluorescence microscope E1000 (Nikon, Tokyo).

Chemotactic assay. Chemotaxis of CD45 primary neuroblasts was tested using 24 transwell plates (5 μm pore size, polycarbonate membrane; Costar, Cambridge, MA). Neuroblasts (2 × 105) were dispensed in the upper chamber and CXCL12 (R&D Systems, Minneapolis, MN; range 30-1,000 ng/mL) or medium alone was added to the lower chamber. Paired CD45+ cell fractions and the Tanque cell line were used as positive controls. Plates were incubated for 2 hours at 37°C. In some experiments, incubation time was prolonged to 4 or 6 hours. After removal of the transwell insert, cells from the lower compartments were collected, counted by trypan blue staining, and analyzed for GD2 and CD45 surface expression. Chemotaxis of SH-SY-5Y, HTLA-230, IMR-32, LAN-5, and GI-LI-N to CXCL12 was also tested using 8 or 12 μm pore-size, polycarbonate membranes. In some experiments, the incubation time of cell lines was prolonged to 4, 6, or 16 hours.

Actin polymerization. Actin polymerization was investigated as previously described (22). Briefly, neuroblastoma cell lines (1 × 106 cells/mL) or the BRG control cell line were incubated at 37°C in RPMI medium containing 20 mmol/L HEPES (Sigma) with or without different concentrations of CXCL12 (50-1,000 ng/mL). After 20 seconds, 100 μL of cell suspension was added to 400 μL of 10−7 mol/L Alexa 488-labeled phalloidin (Molecular Probes, Invitrogen, Carlsbad, CA), 0.125 mg/mL l-α-lysophosphatidylcholine and 4.5% formaldehyde (PLF mix) in PBS (Sigma). Fixed cells were then analyzed by flow cytometry. In another set of experiments, every 15 seconds, 100 μL of cell suspension, incubated with or without CXCL12 (100 ng/mL), was added to the PLF mix before flow cytometric analysis.

Phosphorylated MAPK detection by flow cytometry. The phosphorylation of MAPK following CXCL12 stimulation of neuroblastoma cell lines was evaluated by flow cytometry as previously described (27). Briefly, neuroblastoma or BRG cells were serum-starved for 18 hours before being incubated for 5 minutes with human CXCL12 (100 or 500 ng/mL). Subsequently, cells were fixed in 1% paraformaldehyde and permeabilized with ice-cold methanol (Sigma). Cells (1 × 106) were then incubated with 20 μL of FITC-labeled rabbit anti-MAPK (pTpY185/187) in 2% PBS, 0.1% saponin (Sigma), FCS for 30 minutes. After extensive washing, cells were resuspended in 500 μL of PBS before being analyzed by flow cytometry.

Reverse transcription-PCR. RNA was extracted from primary neuroblasts or neuroblastoma cell lines using RNeasy mini kit from Qiagen GmbH (Hilden, Germany) and subjected to RT-PCR. Primer sequences and amplification profiles were the following: G3PDH, 5′ ACATCGCTCAGAACACCTATGG and 3′ GGGTCTACATGGCAACTGTGAG; CXCR4, 5′ ATCTTCTCGCCCACCATC and 3′ CACCTCGCTTTCCTTTGG; RGS1, 5′ TCTCTGCTAACCCAAAGG and 3′ ACAAGCCAGCCAGAACTC; RGS3, 5′ CAGAAGGCAGGGGCAGAG and 3′ TGAGGTGGGCTTGAATGACT; RGS13, 5′ AACATTGCTTTTCTTTTCTC and 3′ TGTGGAGGGTAGGTTGAGTG; RGS16, 5′ GGAGCAACAGGCGTGGAG and 3′ AACAGTCCAACCAACAACCT; RGS18, 5′ TGGAATGGATATGAAAGC and 3′ TTGATTTGTTTAGCAGCA. Amplification profile was 94°C for 1 minute, annealing at 60°C (G3PDH), 55°C (CXCR4), 49°C (RGS1, RGS13, and RGS16), 51°C (RGS16), 58°C (RGS18) for 1 minute and extension at 72°C for 1 minute. Each cycle of amplification was repeated 35 times. Ten microliters of each sample were electrophoresed through a 1% agarose gel containing ethidium bromide. The specificity of amplification products was checked by confirming the known base-pair sequence length.

Short interfering RNA and RNA interference. LAN-5, GI-LI-N, IMR-32, and SH-SY-5Y neuroblastoma cell lines were transfected with RGS16 or unrelated short interfering RNA (siRNA) purchased from Ambion (Austin, TX). Three different short interfering RGS16 sequences were used with similar results in terms of down-regulation of RGS16 transcript. The sequences are the following: RGS16 A, 5′ GGAUCCUAGGAAAUAAGUCtt and 3′ GACUUAUUUCCUAGGAUCCtt; RGS16 B, 5′ GGAAGUGUUCUAGUGUGGUtt and 3′ ACCACACUAGAACACUUCCtt; RGS16 C, 5′ GGGCACACCAGAUCUUUGAtt and 3′ UCAAAGAUCUGGUGUGCCCtg. Transfection was carried out using 3 × 105 cells, the siPORT NeoFX transfection agent (Ambion), 150 nmol/L of siRNA and following the guidelines of the manufacturer.

Quantitative RT-PCR. The cell lines were subjected to RNA extraction using the RNeasy micro kit (Qiagen), 24, 48 and 72 hours after transfection with the RGS16 or unrelated siRNA. One microgram of total RNA was reverse-transcribed in a total volume of 30 μL. Five microliters of cDNA were amplified in triplicate samples using primers and probes specific for β2-microglobulin or for RGS16 (Applied Biosystems, Foster City, CA). To estimate the relative amount of RGS16 transcript, the comparative threshold method (28) was employed using β2-microglobulin as endogenous reference and the amount of RGS16 transcript in each untransfected cell line as calibrator.

CXCR4 expression and chemotaxis of primary neuroblastoma cells in response to CXCL12. Bone marrow aspirates from six patients with stage IV neuroblastoma were found to contain GD2+ neuroblasts in the following proportions: BM1, 34.3%; BM2, 31.6%; BM3, 50.8%; BM4, 25.6%; BM5, 65.7%; and BM6, 44.4%. GD2+ neuroblasts coexpressing CXCR4 were detected in all bone marrow samples and ranged from 25% to 85% (mean, 55%). Two representative experiments carried out with BM1 and BM5 are shown in Fig. 1A.

Fig. 1.

CXCR4 expression and chemotaxis of primary neuroblastoma cells in response to CXCL12. A, representative CXCR4 surface staining of GD2+ neuroblasts from two bone marrow aspirates (BM1 and BM5). Dark profiles, cells stained with an IgG2a antibody of irrelevant specificity (anti-GD2 control), followed by phycoerythrin-conjugated goat anti-mouse IgG2a antibodies. After washing, cells were incubated with an IgG1 antibody of irrelevant specificity (anti-CXCR4 control), followed by FITC-conjugated goat anti-mouse IgG1 antibodies. Open profile, staining obtained using anti-GD2 mAb, followed by phycoerythrin-conjugated goat anti-mouse IgG2a antibodies, an anti-CXCR4 mAb and, subsequently, FITC-conjugated goat anti-mouse IgG1 antibodies. The profiles were obtained by setting the gate on GD2+ cells. B, chemotaxis of neuroblasts isolated from bone marrow aspirates in response to CXCL12. CD45 neuroblasts and CD45+ cell fractions were incubated with 300 ng/mL of CXCL12 or medium for 2 hours in transwell plates. Results are means from experiments done using six different bone marrow fractions.

Fig. 1.

CXCR4 expression and chemotaxis of primary neuroblastoma cells in response to CXCL12. A, representative CXCR4 surface staining of GD2+ neuroblasts from two bone marrow aspirates (BM1 and BM5). Dark profiles, cells stained with an IgG2a antibody of irrelevant specificity (anti-GD2 control), followed by phycoerythrin-conjugated goat anti-mouse IgG2a antibodies. After washing, cells were incubated with an IgG1 antibody of irrelevant specificity (anti-CXCR4 control), followed by FITC-conjugated goat anti-mouse IgG1 antibodies. Open profile, staining obtained using anti-GD2 mAb, followed by phycoerythrin-conjugated goat anti-mouse IgG2a antibodies, an anti-CXCR4 mAb and, subsequently, FITC-conjugated goat anti-mouse IgG1 antibodies. The profiles were obtained by setting the gate on GD2+ cells. B, chemotaxis of neuroblasts isolated from bone marrow aspirates in response to CXCL12. CD45 neuroblasts and CD45+ cell fractions were incubated with 300 ng/mL of CXCL12 or medium for 2 hours in transwell plates. Results are means from experiments done using six different bone marrow fractions.

Close modal

The CD45 cell fraction isolated from bone marrow samples contained >90% neuroblasts, as assessed by staining with anti-GD2 mAb; in contrast, <5% GD2+ neuroblasts were consistently detected in purified CD45+ cells.

Next, chemotaxis of paired CD45 and CD45+ bone marrow cell fractions to CXCL12 was investigated. CD45 neuroblasts did not migrate to a wide range of CXCL12 concentrations (30, 100, 300, 500, and 1,000 ng/mL). In contrast, the CD45+ cell fractions showed a sustained chemotactic response to 100 ng/mL CXCL12 (Fig. 1B).

CXCR4 expression and chemotaxis of neuroblastoma cell lines in response to CXCL12. As shown in Fig. 2A, CXCR4 mRNA was found to be expressed in all neuroblastoma cell lines. Surface CXCR4 was detected on GI-LI-N (34% positive cells), LAN-5 (30% positive cells), IMR-32 (23% positive cells), HTLA-230 (16% positive cells), and SH-SY-5Y (5.5% positive cells), but not in ACN, GI-ME-N, GI-CA-N, LAN-1, SK-N-SH, or SK-N-BE-2c, neuroblastoma cell lines (Fig. 2B). These figures were stable over time, as shown by repeated testing of cell lines at 2-month intervals.

Fig. 2.

CXCR4 expression, migration, and actin modulation in neuroblastoma cell lines exposed to CXCL12. A, CXCR4 expression in neuroblastoma cell lines by RT-PCR. From left to right: MW, molecular weight marker; NC, negative control represented by water in the place of cDNA; PC, positive control represented by mononuclear cells isolated from tonsil; lanes 4 to 13, 10 neuroblastoma cell lines. Amplification product of the G3PDH housekeeping gene tested as control (top). B, CXCR4 expression on neuroblastoma cell lines cultured in normoxic or hypoxic conditions, as assessed by flow cytometry. Three representative neuroblastoma cell lines are shown. Dark profile in the three panels were obtained by staining neuroblastoma cells cultured in normoxic condition with an IgG1 antibody of irrelevant specificity (anti-CXCR4 control), followed by FITC-conjugated goat anti-mouse IgG1 antibodies. Superimposable results were obtained with the same staining of neuroblastoma cells cultured in hypoxic conditions. CXCR4 staining in IMR-32 (left), HTLA-230 (middle), and LAN-5 (right) maintained in normoxic (open profile) or in hypoxic condition (broken line). C, neuroblastoma cell line migration in response to CXCL12 (300 ng/mL) maintained under normoxic (black columns) or hypoxic condition (gray columns); migration of cells in the presence of medium alone (open columns). The Tanque cell line was used as a positive control. D, actin modulation in HLTA-230 neuroblastoma cell line in response to different concentration of CXCL12, as assessed by flow cytometry. As control, the BRG cell line was treated for different time intervals with a fixed concentration of CXCL12 (100 ng/mL).

Fig. 2.

CXCR4 expression, migration, and actin modulation in neuroblastoma cell lines exposed to CXCL12. A, CXCR4 expression in neuroblastoma cell lines by RT-PCR. From left to right: MW, molecular weight marker; NC, negative control represented by water in the place of cDNA; PC, positive control represented by mononuclear cells isolated from tonsil; lanes 4 to 13, 10 neuroblastoma cell lines. Amplification product of the G3PDH housekeeping gene tested as control (top). B, CXCR4 expression on neuroblastoma cell lines cultured in normoxic or hypoxic conditions, as assessed by flow cytometry. Three representative neuroblastoma cell lines are shown. Dark profile in the three panels were obtained by staining neuroblastoma cells cultured in normoxic condition with an IgG1 antibody of irrelevant specificity (anti-CXCR4 control), followed by FITC-conjugated goat anti-mouse IgG1 antibodies. Superimposable results were obtained with the same staining of neuroblastoma cells cultured in hypoxic conditions. CXCR4 staining in IMR-32 (left), HTLA-230 (middle), and LAN-5 (right) maintained in normoxic (open profile) or in hypoxic condition (broken line). C, neuroblastoma cell line migration in response to CXCL12 (300 ng/mL) maintained under normoxic (black columns) or hypoxic condition (gray columns); migration of cells in the presence of medium alone (open columns). The Tanque cell line was used as a positive control. D, actin modulation in HLTA-230 neuroblastoma cell line in response to different concentration of CXCL12, as assessed by flow cytometry. As control, the BRG cell line was treated for different time intervals with a fixed concentration of CXCL12 (100 ng/mL).

Close modal

The seemingly low proportions of CXCR4+ SH-SY-5Y cells here detected in comparison with previous data [see Sanders et al. in ref. (9)] prompted additional experiments in which CXCR4 expression was investigated in this cell line by immunocytochemistry. Indeed, CXCR4 was found to be expressed at very low intensities in the majority of cells and at higher intensities in the minority of cells (5-10%). A likely explanation for these apparently discrepant findings is the usage in this study of cells that have somewhat down-regulated CXCR4 due to considerable time in culture.

Following culture of neuroblastoma cells in 1% oxygen tension, which mimics the hypoxic conditions found in the tumor microenvironment, a strong CXCR4 up-regulation was observed (Fig. 2B). The proportions of CXCR4+ cells raised to 92% in GI-LI-N cells, 97% in LAN-5 cells, 93% in IMR-32 cells, 85% in HTLA-230 cells, and 40% in SH-SY-5Y cells.

The five CXCR4+ neuroblastoma cell lines did not migrate in a transwell assay to CXCL12 when tested at different concentrations (30, 100, 300, 500, and 1,000 ng/mL) or at different times (2, 4, 6, and 16 hours; Fig. 2C). Likewise, no chemotaxis was detected when the pore size of the filter was increased from 5 to 8 or 12 μm. In contrast, in the same experiments, the Tanque cell line (control) was attracted by CXCL12.

Next, these experiments were repeated in the anaerobic work station following overnight preincubation of cell lines in 1% O2. Again no chemotaxis of CXCR4+ neuroblastoma cell lines were observed at any CXCL12 concentration, whereas Tanque cells migrated to CXCL12 (Fig. 2C). In addition, CD45 bone marrow neuroblasts failed to migrate to CXCL12 in the same experimental conditions, whereas paired CD45+ bone marrow cell fractions were attracted by the chemokine (data not shown).

Redistribution of actin filaments (F-actin) on the cell surface is an early event associated with chemotaxis. We therefore investigated F-actin modulation in the HTLA-230 cell line following incubation with increasing concentrations of CXCL12. As shown in Fig. 2D (left), 20 seconds of incubation of neuroblastoma cells with CXCL12 did not modify actin polymerization at any concentration tested, as assessed by flow cytometry. The same result was obtained when F-actin modulation was investigated at different time points. Figure 2D (right) shows time-dependent F-actin modulation detected in BRG cells upon incubation with CXCL12.

Expression of RGS in bone marrow neuroblasts and in neuroblastoma cell lines. To gain insight into the potential mechanism(s) underlying the chemotactic unresponsiveness of neuroblasts to CXCL12, expression of a panel of RGS proteins involved in the negative control of CXCR4 signaling (RGS1, RGS3, RGS13, RGS16, and RGS18) was next investigated by RT-PCR. These experiments were carried out with primary neuroblasts isolated from four bone marrow samples and with four CXCR4+ neuroblastoma cell lines. Figure 3A shows the results obtained with these cell lines and two representative bone marrow samples.

Fig. 3.

RGS expression in neuroblastoma cell lines and in neuroblasts from bone marrow aspirates, MAPK activation in HTLA-230 cells and down-regulation of RGS16 by siRNA in neuroblastoma cell lines. A, RGS expression in CXCR4+ neuroblastoma cell lines, tested by RT-PCR. MW, molecular weight marker; NC, negative control represented by water in the place of cDNA; PC, positive control represented by mononuclear cells isolated from the tonsil; lanes 4 to 7, HTLA-230, LAN-5, SK-SY-5Y, and IMR-32 are shown; lanes 8 and 9, neuroblastoma (CD45) cells purified from two representative bone marrow aspirates (patients 1 and 2). B, phosphorylation of MAPK in HTLA-230 cell line incubated with medium alone (dark profile), 100 (thin line), or 500 ng/mL (thick line) CXCL12. Cells were stained with a FITC-labeled rabbit anti-MAPK (pTpY185/187) in 2% PBS, 0.1% saponin (Sigma), FCS for 30 minutes, and subsequently analyzed by flow cytometry. As control, the BRG cell line was treated with 100 ng/mL CXCL12. C, down-regulation of RGS16 transcript in GI-LI-N, IMR-32, LAN-5, and SH-SY5Y cells, either transfected with RGS16-specific siRNA (gray columns) or unrelated siRNA (white columns), or untransfected (no siRNA, black columns). Inhibition of RGS16 mRNA was investigated by quantitative RT-PCR.

Fig. 3.

RGS expression in neuroblastoma cell lines and in neuroblasts from bone marrow aspirates, MAPK activation in HTLA-230 cells and down-regulation of RGS16 by siRNA in neuroblastoma cell lines. A, RGS expression in CXCR4+ neuroblastoma cell lines, tested by RT-PCR. MW, molecular weight marker; NC, negative control represented by water in the place of cDNA; PC, positive control represented by mononuclear cells isolated from the tonsil; lanes 4 to 7, HTLA-230, LAN-5, SK-SY-5Y, and IMR-32 are shown; lanes 8 and 9, neuroblastoma (CD45) cells purified from two representative bone marrow aspirates (patients 1 and 2). B, phosphorylation of MAPK in HTLA-230 cell line incubated with medium alone (dark profile), 100 (thin line), or 500 ng/mL (thick line) CXCL12. Cells were stained with a FITC-labeled rabbit anti-MAPK (pTpY185/187) in 2% PBS, 0.1% saponin (Sigma), FCS for 30 minutes, and subsequently analyzed by flow cytometry. As control, the BRG cell line was treated with 100 ng/mL CXCL12. C, down-regulation of RGS16 transcript in GI-LI-N, IMR-32, LAN-5, and SH-SY5Y cells, either transfected with RGS16-specific siRNA (gray columns) or unrelated siRNA (white columns), or untransfected (no siRNA, black columns). Inhibition of RGS16 mRNA was investigated by quantitative RT-PCR.

Close modal

RGS1, RGS3, RGS13, RGS16, and RGS18 transcripts were consistently expressed in primary CD45 neuroblasts (Fig. 3A). RGS16 mRNA was detected in all neuroblastoma cell lines, RGS3 transcript was found in HTLA-230, LAN-5, and SK-SY-5Y, but not in IMR-32 cells, and RGS13 mRNA was expressed in LAN-5 cells only. RGS1 and RGS18 mRNAs were never detected in any cell line (Fig. 3A).

RGS3 and RGS16 interfere with CXCR4 signal transduction by reducing the duration and magnitude of G protein signaling, and thus at least in part inhibiting MAPK phosphorylation (1518). Therefore, using flow cytometry, we investigated MAPK phosphorylation in the HTLA-230 neuroblastoma cell line following treatment with CXCL12. As shown in Fig. 3B (left), phosphorylated MAPK was not detected under these conditions, suggesting that CXCR4-dependent MAPK activation was impaired. In contrast, BRG cells incubated with CXCL12 displayed evident MAPK activation (Fig. 3B, right).

Because RGS16 was always detected in primary neuroblasts and neuroblastoma cell lines, we next addressed its potential role in dampening CXCR4 signaling. RGS16 protein expression, as assessed by immunofluorescence was found to be low in LAN-5, GI-LI-N, IMR-32, and SH-SY-5Y cells (range, 3-10%) and was absent in HTLA-230 cells.

Next, LAN-5, GI-LI-N, IMR-32, and SH-SY-5Y cells were transfected with RGS16-specific or unrelated siRNA, and inhibition of RGS16 transcription was investigated by quantitative RT-PCR. As shown in Fig. 3C, RGS16 mRNA levels were strongly down-regulated 24 hours after transfection with RGS16-specific, but not in unrelated, siRNA. This effect was no longer detectable after 48 or 72 hours from transfection (data not shown). The low constitutive expression of the RGS16 protein did not allow us to establish whether the latter had been down-regulated by siRNA transfection.

Finally, LAN-5, GI-LI-N, IMR-32, and SH-SY-5Y cells were subjected to chemotactic assays in the presence or absence of CXCL12 at 24, 48, and 72 hours after transfection. These cell lines were not attracted by CXCL12 at any time tested, suggesting that RGS16 was not involved in CXCR4 functional inactivation (data not shown).

Various in vivo studies have shown that CXCR4 inhibition by antibodies or small antagonists counteracts the metastatic potential of tumor cells (13, 29). Thus, a new promising area of clinical investigation aimed at targeting metastatic tumor cells is emerging; examples of malignancies potentially amenable to such treatment are breast, ovarian, pancreatic, and prostate cancers, gliomas, as well as different leukemias and lymphomas (13).

Previously, it was shown that a panel of neuroblastoma cell lines expressed CXCR4 at low levels and that SH-SY5Y did not migrate to CXCL12 (9). However, following transfection of the CXCR4 gene and consequent up-regulation of surface CXCR4, the latter cells were attracted by CXCL12 and invaded in vitro CXCL12 producing bone marrow stromal cells (9), suggesting CXCR4 involvement in bone marrow metastasis formation by primary neuroblastoma cells.

In this study, we show unambiguously that the majority of human metastatic neuroblastoma cells isolated from the bone marrow express CXCR4 but are not attracted by the CXCR4-specific ligand, CXCL12. Similar results were obtained with five CXCR4+ neuroblastoma cell lines. Because the tumor microenvironment is hypoxic, we investigated whether exposure of neuroblastoma cells to low oxygen tension influenced CXCR4 expression and functionality. Although CXCR4 was strongly up-regulated under the latter conditions, no tumor cell chemotaxis was ever detected over a wide range of CXCL12 concentrations. Accordingly, neuroblastoma cells did not modulate F-actin nor activated MAPK upon incubation with CXCL12.

Taken together, these result do not support the hypothesis that the CXCR4-CXCL12 axis is involved in the metastatic localization of human neuroblastoma cells to the bone marrow. However, the possibility that primary CXCR4+ neuroblastoma cells are attracted to the bone marrow by a CXCL12-dependent gradient and subsequently undergo ligand-dependent CXCR4 desensitization cannot be ruled out.

The role of chemokine networks in tumor spreading goes beyond directing the movement of malignant cells. Thus, CXCL12 produced by neuroblastoma cell lines stimulated their growth through autocrine and/or paracrine loops (9), and CXCR4 inhibition with AMD3100 induced apoptosis of neuroblastoma cells in vitro (30). CXCR4-mediated chemotaxis and proliferation triggered by CXCL12 could depend on different and mutually exclusive signaling pathways (8, 31, 32), leaving open the possibility that CXCR4 inhibitors could be used in the clinical setting to interfere with neuroblastoma cell growth rather than with metastatic spreading.

All primary neuroblastoma cell samples and neuroblastoma cell lines were found to express RGS16 transcript, pointing to the potential involvement of the RGS16 protein in CXCR4 inactivation. This hypothesis was tested by down-regulating RGS16 mRNA in CXCR4+ cell lines through RNA interference. However, no CXCL12-driven chemotaxis of neuroblastoma cells was observed under these conditions. These results suggest that other still undefined mechanisms are responsible for chemotactic inactivation of CXCR4 in human neuroblastoma cells.

Grant support: Ministero dell'Istruzione, Università e Ricerca and Compagnia San Paolo di Torino (V. Pistoria), Fondaziona Italiana per la Ricerca sul Cancro fellowship (C. Cocco), and Fondazione Italiana per la Lotta al Neuroblastoma fellowship (B. Carlini).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: I. Airoldi and L. Raffaghello contributed equally to this work.

M.V. Corrias and V. Pistoia contributed equally to this work.

We thank Chiara Bernardini for excellent secretarial assistance.

1
Scotton CJ, Wilson JL, Scott K, et al. Multiple actions of the chemokine CXCL12 on epithelial tumor cells in human ovarian cancer.
Cancer Res
2002
;
62
:
5930
–8.
2
Scala S, Ottaiano A, Ascierto PA, et al. Expression of CXCR4 predicts poor prognosis in patients with malignant melanoma.
Clin Cancer Res
2005
;
11
:
1835
–41.
3
Marchesi F, Monti P, Leone BE, et al. Increased survival, proliferation, and migration in metastatic human pancreatic tumor cells expressing functional CXCR4.
Cancer Res
2004
;
64
:
8420
–27.
4
Smith MC, Luker KE, Garbow JR, et al. CXCR4 regulates growth of both primary and metastatic breast cancer.
Cancer Res
2004
;
64
:
8604
–12.
5
Russell HV, Hicks J, Okcu MF, Nuchtern JG. CXCR4 expression in neuroblastoma primary tumors is associated with clinical presentation of bone and bone marrow metastases.
J Pediatr Surg
2004
;
39
:
1506
–11.
6
Salmaggi A, Gelati M, Pollo B, et al. M. CXCL12 in malignant glial tumors: a possible role in angiogenesis and cross-talk between endothelial and tumoral cells.
J Neurooncol
2004
;
67
:
305
–17.
7
Burger M, Glodek A, Hartmann T, et al. Functional expression of CXCR4 (CD184) on small-cell lung cancer cells mediates migration, integrin activation, and adhesion to stromal cells.
Oncogene
2003
;
22
:
8093
–101.
8
Hwang JH, Chung HK, Kim DW, et al. CXC chemokine receptor 4 expression and function in human anaplastic thyroid cancer cells.
J Clin Endocrinol Metab
2003
;
88
:
408
–16.
9
Geminder H, Sagi-Assif O, Goldberg L, et al. A possible role for CXCR4 and its ligand, the CXC chemokine stromal cell-derived factor-1, in the development of bone marrow metastases in neuroblastoma.
J Immunol
2001
;
167
:
4747
–57.
10
Corcione A, Ottonello L, Tortolina G, et al. Stromal cell-derived factor-1 as a chemoattractant for follicular center lymphoma B cells.
J Natl Cancer Inst
2000
;
92
:
628
–35.
11
Schimanski CC, Schwald S, Simiantonaki N, et al. Effect of chemokine receptors CXCR4 and CCR7 on the metastatic behavior of human colorectal cancer.
Clin Cancer Res
2005
;
11
:
1743
–50.
12
Sun YX, Wang J, Shelburne CE, et al. Expression of CXCR4 and CXCL12 (SDF-1) in human prostate cancers (PCa) in vivo.
J Cell Biochem
2003
;
89
:
462
–73.
13
Balkwill F. Cancer and the chemokine network.
Nat Rev Cancer
2004
;
4
:
540
–50.
14
Hollinger S, Hepler JR. Cellular regulation of RGS proteins: modulators and integrators of G protein signaling.
Pharmacol Rev
2002
;
54
:
527
–59.
15
Johnson EN, Druey KM. Functional characterization of the G protein regulator RGS13.
J Biol Chem
2002
;
277
:
16768
–74.
16
Shi CS, Lee SB, Sinnarajah S, Dessauer CW, Rhee SG, Kehrl JH. Regulator of G-protein signaling 3 (RGS3) inhibits Gβ1γ 2-induced inositol phosphate production, mitogen-activated protein kinase activation, and Akt activation.
J Biol Chem
2001
;
276
:
24293
–300.
17
Zhang Y, Neo SY, Han J, Yaw LP, Lin SC. RGS16 attenuates gαq-dependent p38 mitogen-activated protein kinase activation by platelet-activating factor.
J Biol Chem
1999
;
274
:
2851
–7.
18
Druey KM, Blumer KJ, Kang VH, Kehrl JH. Inhibition of G-protein-mediated MAP kinase activation by a new mammalian gene family.
Nature
1996
;
379
:
742
–6.
19
Shi GX, Harrison K, Han SB, Moratz C, Kehrl JH. Toll-like receptor signaling alters the expression of regulator of G protein signaling proteins in dendritic cells: implications for G protein-coupled receptor signaling.
J Immunol
2004
;
172
:
5175
–84.
20
Shi GX, Harrison K, Wilson GL, Moratz C, Kehrl JH. RGS13 regulates germinal center B lymphocytes responsiveness to CXC chemokine ligand (CXCL)12 and CXCL13.
J Immunol
2002
;
169
:
2507
–15.
21
Lippert E, Yowe DL, Gonzalo JA, et al. Role of regulator of G protein signaling 16 in inflammation-induced T lymphocyte migration and activation.
J Immunol
2003
;
171
:
1542
–55.
22
Piovan E, Tosello V, Indraccolo S, et al. Chemokine receptor expression in EBV-associated lymphoproliferation in hu/SCID mice: implications for CXCL12/CXCR4 axis in lymphoma generation.
Blood
2005
;
105
:
931
–9.
23
Cotterill SJ, Pearson AD, Pritchard J, Kohler JA, Foot AB. Late relapse and prognosis for neuroblastoma patients surviving 5 years or more: a report from the European Neuroblastoma Study Group “Survey”.
Med Pediatr Oncol
2001
;
36
:
235
–8.
24
Brodeur GM, Seeger RC, Barrett A, et al. International criteria for diagnosis, staging and response to treatment in patients with neuroblastoma.
Prog Clin Biol Res
1988
;
271
:
509
–24.
25
Puppo M, Pastorino S, Melillo G, Pezzolo A, Varesio L, Bosco MC. Induction of apoptosis by flavopiridol in human neuroblastoma cells is enhanced under hypoxia and associated with N-myc proto-oncogene down-regulation.
Clin Cancer Res
2004
;
10
:
8704
–19.
26
Airoldi I, Gri G, Marshall JD, et al. Expression and function of IL-12 and IL-18 receptors on human tonsillar B cells.
J Immunol
2000
;
165
:
6880
–8.
27
Fleisher AS, Esteller M, Wang S, et al. Hypermethylation of the hMLH1 gene promoter in human gastric cancers with microsatellite instability.
Cancer Res
1999
;
59
:
1090
–95.
28
Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Δ ΔC(T)) Method.
Methods
2001
;
25
:
402
–8.
29
Muller A, Homey B, Soto H, et al. Involvement of chemokine receptors in breast cancer metastasis.
Nature
2001
;
410
:
50
–6.
30
Hatse S, Bridger G, de Clercq E, Schols D. X4 HIV-1 induces neuroblastoma cell death by interference with CXCL12/CXCR4 interaction.
Cell Mol Biol (Noisy-le-grand)
2003
;
49
Online Pub:
OL443
–52.
31
Bendall LJ, Baraz R, Juarez J, Shen W, Bradstock KF. Defective p38 mitogen-activated protein kinase signaling impairs chemotaxic but not proliferative responses to stromal-derived factor-1α in acute lymphoblastic leukemia.
Cancer Res
2005
;
65
:
3290
–8.
32
Ahr B, Denizot M, Robert-Hebmann V, Brelot A, Biard-Piechaczyk M. Identification of the cytoplasmic domains of CXCR4 involved in Jak2 and STAT3 phosphorylation.
J Biol Chem
2005
;
280
:
6692
–700.