SEARCH

SEARCH BY CITATION

Keywords:

  • neuroblastoma;
  • galectin-1;
  • tolerance;
  • T cells;
  • dendritic cells

Abstract

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The immunosuppressive strategies devised by neuroblastoma (NB), the most common solid extracranial childhood cancer, are poorly understood. Here, we identified an immunoevasive program triggered by NB through secretion of galectin-1 (Gal-1), a multifunctional glycan-binding protein. Human and mouse NB cells express and secrete Gal-1, which negatively regulates T cell and dendritic cell function. When injected subcutaneously in syngeneic A/J mice, knockdown transfectants expressing low amounts of Gal-1 (NXS2/L) showed reduction of primary tumor growth by 83–90% and prevented spontaneous liver metastases in contrast to NXS2 cell variants (NXS2/H, NXS2 wildtype) expressing high amounts of Gal-1. Splenocytes from mice receiving Gal-1 knockdown NXS2/L cells secreted higher amounts of IFN-γ and displayed enhanced cytotoxic T-cell function compared to NXS2/H or NXS2 controls. Immunohistochemical analysis revealed a six- to tenfold increase in the frequency of CD4+ and CD8+ T cells infiltrating tumors from mice receiving knockdown transfectants. This effect was confirmed by in vitro migration assays. Finally, supernatants of NXS2/H or NXS2 cells suppressed dendritic cell (DC) maturation and induce T cell apoptosis, whereas these effects were only marginal on DCs and T cells exposed to supernatants from NXS2/L cells. These results demonstrate a novel immunoinhibitory role of the Gal-1-glycan axis in NB, highlighting an alternative target for novel immunotherapeutic modalities.

Neuroblastoma (NB) is the most common extracranial solid malignancy in childhood.1 This neuroendocrine tumor arising from neural crest cells is responsible for over 10–15% of pediatric cancer deaths.1 The disease exhibits extreme heterogeneity, resulting in most recent stratification into very low, low, intermediate or high risk.2 NB of lower risk groups occur during the first 18 months of life, and good outcomes are typical after surgery.2 In contrast, high-risk NB, notably characterized by the amplification of NB-derived myelocytomatosis viral related oncogene (N-MYC), is resistant even to the most intensive treatment protocols available.3, 4 This suggests the need of more effective therapeutic strategies targeting nontraditional targets and lacking toxicity for treating high-risk aggressive NB.

The introduction of monoclonal antibodies or chimeric T cells engineered to express chimeric antigen receptors targeting the GD2 antigen together with the implementation of dendritic cell (DC)-based vaccination have validated the critical function of the immune system in promoting NB regression.5 Treatment with IL-2, GM-CSF, CD40 agonists or CTLA-4 blockade, in combination with other traditional approaches, has successfully improved immune effector functions in high-risk NB patient.5, 6 However, the success of these adjuvant immunotherapeutic approaches is hindered by a number of strategies used by tumors or tumor-associated stroma cells to elude immune recognition or thwart immune responses.7 Although well defined in a number of tumor types (e.g. melanoma, ovary cancer and prostate carcinoma),7 the immunoevasive programs displayed by neuroendocrine tumors, particularly NB cells, are still poorly understood.

Galectins, a family of evolutionarily conserved glycan-binding proteins, can control different steps of tumor progression and metastasis including cellular transformation, adhesion, migration, immunity and angiogenesis.8–10 Intracellularly, galectins interact with signaling pathways, whereas extracellularly, these soluble proteins function by cross-linking cell surface glycoconjugates, forming multimeric glycan arrays termed “lattices” and modulating intracellular signalling pathways.11 Research over the past years using experimental models of autoimmunity, inflammation, pregnancy and cancer, has provided proof-of-concept of the pivotal role of galectin-1 (Gal-1), a “proto-type” member of the galectin family, in immune tolerance and homeostasis. In fact, tumor-secreted Gal-1 contributes to the immunosuppressive activity of melanoma,12 Hodgkin lymphoma,13 lung carcinoma14, 15 and pancreatic carcinoma.16 Targeted disruption of Gal-1 gene expression resulted in enhanced T cell-mediated tumor rejection,12 suggesting novel therapeutic opportunities for immunointervention based on the selective blockade of Gal-1-glycan interactions. The cellular and molecular mechanisms underlying these immunosuppressive activities have been recently examined, demonstrating that Gal-1 selectively controls Th1- and Th17-mediated effector functions17 and induces the differentiation of tolerogenic dendritic cells (DCs) through IL-27- and IL-10-dependent pathways.18

Tumor cells from the central nervous system, such as malignant glioma cells, express high amounts of Gal-1, which plays multifaceted roles including modulation of the migratory and invasive capacity of cancer cells, induction of angiogenesis and resistance to chemotherapy.19, 20 Interestingly, NB tumors, particularly those with poor prognosis, also express high levels of Gal-1.21 Blockade of intracellular Gal-1 reduced the migratory and invasive properties of TrkB+ NB cancer cells.21 Moreover, recombinant Gal-1 (rGal-1) restored NB cell invasiveness in these cell cultures suggesting an extracellular action of this lectin in eliciting glycan-dependent signalling processes.21

Based on these observations, we hypothesized that Gal-1 may be secreted from NB cells to dampen host T cell and DC function, thus making these cells unable to eradicate NB. Here, we show that Gal-1 blockade efficiently inhibits primary tumor growth and metastasis through suppression of T cell cytotoxicity, IFN-γ production and DC maturation in a syngeneic mice model of NB. Our results emphasize the role of Gal-1 as a promising therapeutic target in NB, a paradigmatic aggressive tumor of neuroendocrine origin.

Material and Methods

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Cell culture and viability assay

Murine NXS2, NXS2/H (Gal-1 high), NXS2/L (Gal-1 low) NB cells were cultured in DMEM (Biochrom AG, Berlin, Germany). Human LAN-1, SK-N-SH, SK-N-SA and Kelly, NB cells were cultured in RPMI (Biochrom AG, Berlin, Germany). All media were supplemented with 10% FCS (PAA Laboratories, Pasching, Austria) and 100 μg/ml penicillin-streptomycin (Biochrom AG, Berlin, Germany). Cells were cultured in monolayer at 37°C and 5% CO2. Viability and proliferation of NXS2, NXS2/H and NXS2/L cells were evaluated by cell counting after staining with Trypan blue (Sigma Aldrich, Germany) and using the CellTiter-Glo® Luminescent Cell Viability Assay, performed according to the manufacturer's protocol (Promega, Mannheim, Germany).

Generation of knockdown NB clones

Murine NXS2 NB cells were transfected with the antisense Gal-1 expression vector p6/G1-As12, 22 using Fugene transfection solution (Roche, Mannheim, Germany). Blasticidine-resistant (5 μg/ml, ICN) stable transfectants were cloned by limited dilution, and six clones were generated. Representative clones with low (NXS2/L) or high (NXS2/H) Gal-1 expression were selected for in vitro and in vivo experiments.

DC isolation, culture and maturation experiments

Mouse bone marrow-derived DCs (mBMDCs) were generated from mouse bones cultured for 7 days according to Lutz et al.23 with minor changes. Briefly, bone marrow from mouse femur and tibia were centrifuged (1200 rpm, 5 min, RT) and resuspended in DC media (1.5 × 106/ml; RPMI, 10% FCS, 1% pencillin-streptomycin, 50 μM β-mercaptoethanol (PAA Laboratories, Pasching, Austria)) supplemented with 20 ng/ml GM-CSF (PeproTech, Hamburg, Germany) for 7 day-culture (10 ml/75 cm2 flask, 37°C, 5% CO2) with medium change on day 3 and 5. On day 7, DCs were resuspended in serum-free condition medium (SFCM) (0.3 × 106/ml) generated from 24-h supernatants of confluent NXS2, NXS2/H and NXS2/L cultures simultaneously with LPS (100 ng/ml; Sigma Aldrich, Germany) to induce DC maturation. mBMDCs were also treated with LPS-containing DCM as a control. After 24 h, expression of I-A/I-K, CD86 were analyzed on CD11c+ DCs by flow cytometry.

Flow cytometry

Following a standard protocol with minor changes,24 DCs were carefully washed in PBS (pH 7.4 at 4°C) and resuspended in FACS buffer (PBS, 10% FCS) containing PE-labeled CD11c, FITC-labeled I-A/I-K or CD86 monoclonal antibodies (Becton Dickinson GmbH, Heidelberg, Germany). After washing and fixing in FACS buffer containing 4% paraformaldehyde (Roth, Germany), samples were analyzed using a FACS Calibur or FACS LSR Fortessa (Becton Dickinson, Germany) equipped with Cell Quest and FlowJo.

Cytospins and immunohistochemistry

NB cells were centrifuged in cytospins using microscopic glass slides and after drying fixed in acetone (20°C, 20 min). Analysis of Gal-1 expression was performed using rabbit-monoclonal anti-Gal-1 antibody obtained as described elsewhere22 (1:500, 4°C, overnight) and biotinylated anti- rabbit IgG (1 hr, TRIS buffered saline (TBS), pH 7.5, 2.5% goat serum) after blocking (40 min, TBS, 5% goat serum). Staining was accomplished using ABC reagents (30 min), AEC developing kit (10 min) and hematoxylin (5 min). Antibodies and reagents for immunohistochemistry were purchased from Vector Laboratories (Petersbourgh, UK). T cell infiltrate was analyzed in primary tumors (NXS2) using anti-CD8 monoclonal antibody (clone 53-6.7) and anti-CD4 monoclonal antibody (clone RM4-5) (BD Pharmingen, Heidelberg, Germany) as previously described.24, 25 The average percentage of T cells infiltrating tumors was quantified by light microscopy by analyzing 10 power fields at a magnification of 400×.

Western blot analysis

Supernatants from NXS2, NXS2/H and NXS2/L cells were precipitated with acetone (4:1 volume, 1 hr, −20°C). After centrifugation (4000 rpm, 10 min, 4°C), pellets were dried at RT and resuspended in 0.5× PBS (pH 7.4). Tumor cells and T cells were resuspended in NP-40 lysis buffer (40 min, at 4°C), supplemented with a protease inhibitor (10 mM PMSF Carl Roth GmbH, Co. KG, Karlsruhe, Germany). After centrifugation (12,000 rpm, 30 min, 4°C), pellets were resuspendend in PBS (pH 7.4). Protein concentration was measured using Roti-Nanoquant (Carl Roth GmbH, Co. KG, Karlsruhe, Germany). Denatured proteins (90°C, 5 min) were separated by SDS-PAGE (10 μg per lane) and transferred onto nitrocelullose membranes (Amersham, Buckinghamshire, UK). Blots were incubated with anti-Gal-1 rabbit polyclonal antibody22 or anti-GAPDH monoclonal antibody (Santa Cruz Biotechnology, Inc., Heidelberg, Germany) according to the manufacturer's protocol. To analyze apoptosis, we used anti-active caspase 3 rabbit polyclonal antibody (Abcam, Cambrige, UK) (1:1000, 5% milk, TBS, pH 7.4). HPR-labeled anti-rabbit secondary antibody (Dako GmbH, Hamburg, Germany) was added (1:1000, 5% milk, TBS, pH 7.4). Membranes were washed (3×, 10 min, RT) and revealed with ECL™ Detection Reagents (Amersham, Buckinghamshire, UK).

Mice and tumor challenge experiments

Syngeneic female A/J mice (8–10 weeks old) were used (Harlan Laboratories, Berlin, Germany) and housed in our animal facilities (Otto-von-Guericke University, Medical Faculty, Magdeburg, Germany). All experiments included here were conducted according to the German and NIH guide for the care and use of laboratory animals, i.e. “Tierschutzgesetz des Landes Sachsen-Anhalt.” Protocols were previously approved by the Institutional Review Board of Ministry of Saxony-Anhalt (02-928). The well-established NXS2 model for NB24, 25 was chosen to test our hypothesis. Briefly, primary tumors were induced by injection of NXS2, NXS2/H or NXS2/L NB cells (2 × 106 in 100 μl DMEM, s.c., left flank). Primary tumor growth was analyzed by micro caliper measurements, and tumor volume was calculated as follows (length × width2)/2. When primary tumors of control mice reached an average volume of 300–500 mm3 (after 13–16 days), tumors of all mice were surgically removed and their weights were determined. Tumor specimens were stored at −80°C for further analysis.

Cytotoxicity assay

Splenocytes were isolated from the spleens under sterile conditions and cultured in RPMI (PAA Laboratories, Pasching, Austria) (10% FCS, 100 μg/ml penicillin-streptomycin, 50 μM β-mercaptoethanol, 100 IU/ml IL-2) in the presence of irradiated (30 Gy, 10 min) NXS2 NB cells (100:1) for 7 days. Then, lysis of NXS2-target cells was determined using a standard 51Cr-release assay.24, 25

IFN-γ secretion assay

IFN-γ secretion was measured using a commercial ELISA kit (Becton Dickinson GmbH, Heidelberg, Germany) in the supernatants of splenocytes previously cultured in the presence of irradiated, inactivated NXS2 cells at different time points. These experiments were performed in triplicates.

T cell isolation, migration and annexin V staining

T cells were isolated from the spleens using the Pan T cell isolation kit MACS (Miltenyi Biotec, Bergisch-Gladbach, Germany). Staining and separation were performed according to the manufacturer's protocol. CD3+ T cells were then incubated for 24 hr in NXS2, NXS2/H or NXS2/L SFCM. T cell migration assay was conducted using a two-compartment Boyden chamber system. Isolated CD3+ T cells were loaded into a 5 mm pore polycarbonate transwell insert (Corning Life Science, Amsterdam, Netherlands) (upper chamber, 106 cells in 500 μl RPMI per insert). The lower compartment contained 500 μl SFCM from NXS2, NXS2/H or NXS2/L cells. The negative control was run with RPMI-1640, and the positive control included T cells migrating in the absence of transwell. After 24 hr of incubation, T cells in the lower compartment were counted, and the percentage of T cells migrating (n) was calculated as follows: n/positive control × 100. In another set of experiments, apoptosis of isolated CD3+ T cells was analyzed using a FITC-labeled Annexin V/PI apoptosis detection kit (Bender MedSystems, Vienna, Austria) according to the manufacturer's instructions. Briefly, T cells were washed twice in ice-cold PBS (pH 7.4) and incubated 15 min with FITC-conjugated annexin V and PI staining solution. The percentages of annexin V+ and/or PI+ T cells were analyzed by FACS LSR Fortessa (Becton Dickinson, Germany) equipped with FlowJo.

Data analysis and statistics

Statistical significance between experimental groups was determined using the nonparametrical Kruskal-Wallis test and further analyzed by the Mann-Whitney-U test using Prism 5 software (GraphPad Software, La Jolla, USA). Student's t test was applied to calculate p values for normally distributed data sets. A p value < 0.05 was considered as statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Mouse and human NB cells express and secrete Gal-1

Recent evidence indicates that Gal-1 is highly expressed in SY5Y NB cell lines and in tumor tissue from advanced NB patient.21 To investigate whether Gal-1 confers immune privilege to NB cells, we first confirmed and extended these findings by analyzing Gal-1 expression in both human (LAN-1, Kelly, SK-N-AS and SK-N-SH) and mouse (NXS2) NB cell lines (Fig. 1). Immunohistochemical studies on cytospins of single cell suspensions showed that all NB cells and in particular the murine NXS2 cell line express high intracellular levels of Gal-1 protein (Fig. 1a). Western blot analysis of total cell lysates confirmed these findings, showing considerable expression of this lectin in NXS2, SK-N-AS and SK-N-SH cells, and low expression in Kelly and LAN-1 cells (Fig. 1b). To further investigate whether Gal-1 influences immune responses extracellularly, we examined whether this lectin was released by NB cells. Culture of NB cell lines in SFCM for 24 hr and analysis of Gal-1 secretion revealed high amounts of Gal-1 in supernatants of NXS2, SK-N-AS and SK-N-SH cells but reduced Gal-1 secretion from Kelly and LAN-1 cells (Fig. 1b). Our results indicate that Gal-1 is secreted, although at different extents, by NB tumor cells and can, therefore, act as a soluble factor.

thumbnail image

Figure 1. Expression and secretion of Gal-1 in human and mouse neuroblastoma cell lines. (a) Gal-1 expression was analyzed by immunohistochemistry (IHC) on cytospins (2 × 104 per slide) of murine (NXS2) and human (LAN-1, Kelly and SK-N-SH) NB cell lines at 400× magnification using a specific anti-Gal-1 antibody. Brown staining indicates the presence of Gal-1. NXS2 cells stained without primary antibody were used as negative controls (left panel). (b) Western blot analysis of total cell lysates and supernatants from murine (NXS2) and human (Kelly, LAN-1, SK-N-SA and SK-N-SH) NB cells using a specific anti-Gal-1 antibody. GAPDH were used as loading controls. Fresh culture medium (medium alone) was used as negative control.

Download figure to PowerPoint

Targeted inhibition of Gal-1 gene expression suppresses primary tumor growth and prevents metastases

As Gal-1 expression has been associated with advanced tumor stages, TrkB tyrosine kinase expression and unfavorable prognosis in primary NB,21 we speculated that tumor-secreted Gal-1 may influence NB tumor growth and dissemination in vivo. To address this issue, we transfected murine NXS2 NB cells with Gal-1 antisense cDNA (LAG-1) as described before12 and generated NXS2 clones showing low or high Gal-1 expression (NXS2/L and NXS2/H) for tumor challenge experiments. The expression of Gal-1 in the NXS2/L clone was reduced 80% compared to expression in NXS2/H clone or in wild type nontransfected NXS2 cells (Fig. 2a). Viability and proliferation of NXS2/L and NXS2/H were not substantially different from those showed by the NXS2 (Fig. 2b), confirming that silencing of endogenous Gal-1 did not influence tumor cell growth or apoptosis in vitro. Western blot analysis of total cell lysates from different tissues including muscle, heart, thymus, testis, liver, kidney, lungs, spleen and bone marrow harvested from healthy A/J mice syngeneic to the NXS2 cell clones showed variable, although weak expression of Gal-1 in these tissues, suggesting preferential expression of this lectin in tumor cells in this mouse model (Fig. 2c). To examine whether Gal-1 blockade in the tumor influences its progression, we injected NXS2/L (Gal-1low), NXS2/H or NXS2 (Gal-1high) subcutaneously into A/J mice and assessed primary tumor growth (Fig. 3a). Mice inoculated with the NXS2/L clone showed 83–90% reduction in tumor growth when compared with controls receiving the NXS2/H or NXS2 clones which express high amounts of Gal-1 (Fig. 3b). In addition, weight of surgically removed tumors from NXS2/L-injected mice was 72% lower than tumor weight from mice inoculated with NXS2/H or NXS2 cells (Fig. 3b). In addition, the extent of spontaneous liver metastases was highly dependent on the levels of tumor-specific Gal-1 expression. In this regard, we barely found macroscopic visible metastases in mice receiving NXS2/L transfectants, although metastasis was clearly evident in the NXS2/H group which was comparable to that found in mice receiving NXS2 control tumor cells (Fig. 3c). Thus, endogenous Gal-1 favors tumor growth and metastasis in a syngeneic immunocompetent NB mouse model, suggesting an important role of this lectin in tumor-immune escape.

thumbnail image

Figure 2. Generation and characterization of NXS2 Gal-1 knockdown clones. (a) NXS2/L was generated by stable transfection of Gal-1 antisense cDNA into NXS2 cells. Gal-1 expression was detected by Western blot analysis of total cell lysates of NXS2/L cells (10 μg/lane) using anti-Gal-1 antibody and compared to cell lysates from parental NXS2 and the subclone NXS2/H both expressing high amounts of this lectin. Graph bars show densitometric analysis performed using ImageJ. GAPDH expression was used as control. (b) Cell proliferation was measured using CellTiter-Glo Assay to determine the intracellular level of ATP correlating with viability according to the manufacturer's recommended protocol (left panel). Cell viability was also analyzed after staining of harvested NB cells with Trypan blue (right panel). The percentage of living cells was calculated as follows: (total cell number − number of dead cells)/total cell number. (c) Gal-1 expression on extracts of different mouse tissues (liver, spleen, heart, thymus, muscle, lung, kidney, testis and bone marrow) obtained from A/J mice by Western blot. NXS2 cells were used as positive control.

Download figure to PowerPoint

thumbnail image

Figure 3. Role of endogenous Gal-1 in tumor progression in a syngeneic NB mouse model. (a) Schematic representation of tumor challenge in A/J mice. Mice were challenged s.c. with parental NXS2 (N = 7/group) or its subclones NXS2/L (Gal-1 low, N = 7/group) or NXS2/H (Gal-1 high; N = 4/group) (2 ×106/ 100 μl DMEM) on day 0. On day 13–14, primary tumors were surgically removed, and on day 35, all mice were sacrificed and the occurrence of liver metastases was verified. (b) Tumor growth was determined by caliper measurements (mean ± S.E.M.) over a period of 14 days after induction of primary tumors. The low graphic in b shows the expression of Gal-1 on primary tumors on the day of surgery. Data were analyzed with nonparametric Kruskal-Wallis and further analyzed with Mann-Whitney-U-test; * p ≤ 0.05, ** p ≤ 0.01. (c) The presence of spontaneous liver metastases was determined by the number of metastatic foci in fresh liver specimens. Data were analyzed with nonparametric Kruskal-Wallis and further analyzed with Mann-Whitney-U-test; * p ≤ 0.05, ** p ≤ 0.01. Photographs of one representative liver specimen from each group of mice are shown (bottom panel). Arrows indicate metastatic foci.

Download figure to PowerPoint

Gal-1 impairs recruitment and function of T cells in primary NB

Because T cell subsets involved in tumor immunity share all the repertoire of cell surface glycans critical for Gal-1 binding and signalling,17, 26 we next sought to investigate whether Gal-1 promotes tumor progression through selective inhibition of T-cell effector function. Splenocytes from mice inoculated with NXS2/L tumors showed 20–45% higher cytotoxic activity on target cells (82%; E:T 100:1) when compared with splenocytes from mice receiving the same number of NXS2/H or NXS2 (37%; E:T 100:1) (Fig. 4a). This result suggests a critical role for endogenous Gal-1 in impairing cytotoxic T cell function. In addition, the Th1-type cytokine IFN-γ, which is critical for tumor immunoediting,7 was significantly augmented in splenocyte supernatants from mice receiving NXS2/L tumor cells when compared with those obtained from mice inoculated with NXS2 tumor cells (Fig. 4b). Moreover, Gal-1 released by NB cells influenced migration of T cells in in vitro assays. We found a ∼25–40% increase in the percentage of T cells migrating in response to SFCM from NXS/L (73%) when compared with T cells migrating in response to SFCM from NXS2 (48%) or NXS2/H (33%) tumor cell lines (Fig. 5a). These results suggested that release of Gal-1 from NB cells may contribute to impair migration of effector T cells into the tumor parenchyma. This effect was substantiated by an increased recruitment of CD8+ T and CD4+ T cells into the tumor parenchyma in those mice inoculated with tumors secreting low amounts of Gal-1. In fact, a six- to tenfold increase of both T cell subsets was found in tumors excised from NXS2/L mice when compared with NXS2/H or NXS2 tumors (Fig. 5b). In conclusion, Gal-1 secreted by NB impairs different T cell effector functions, including T-cell-mediated cytotoxicity, IFN-γ secretion and T cell recruitment to tumor parenchyma, indicating a major role for this endogenous lectin in favoring tumor-immune evasion in NB. Blockade of this endogenous lectin significantly contributed to overcome tumor-immune escape by augmenting T cell effector function.

thumbnail image

Figure 4. Targeted inhibition of Gal-1 gene expression increases cytotoxic function and IFN-γ production by T cells. (a) Target cell lysis was determined in triplicates for each group with pooled splenocytes obtained from mice bearing NXS2, NXS2/H (high Gal-1 expression) or NXS2/L (low Gal-1 expression) at varying effector to target cell ratios (1 = 50:1, 2 = 25:1, 3 = 12.5:1). Data are presented as mean ± SEM and were considered to be statistically significant (unpaired t test; * p < 0.05, ** p ≤ 0.01). (b) Supernatants of pooled splenocytes obtained from mice inoculated with NXS2, NXS2/H (high Gal-1 expression) or NXS2/L (low Gal-1 expression) were analyzed for IFN-γ secretion by ELISA (unpaired t test; * p < 0.05, ** p ≤ 0.01).

Download figure to PowerPoint

thumbnail image

Figure 5. Targeting Gal-1 expression augments T cell migration in vitro and favors tumor infiltration by CD4+ and CD8+ T cells. (a) Effect of SFCM collected from NXS2, NXS2/H or NXS2/L cells on the migration of isolated CD3+ T cells into a two-compartment migration system containing a 5 μm transmembrane. Migration is shown as percentage of total T cells migrating relative to those, migrating in the absence of transwell (positive control; 100%). Migration chamber with RPMI-1640 in the lower chamber was used as a negative control. Experiments were performed twice in triplicates and analyzed using unpaired t test; ** p < 0.01, *** p < 0.001. (b) Infiltration of CD4+ (upper panel) and CD8+ (lower panel) T cells in primary tumors is illustrated in scattergrams (NXS2/H, N = 4/group; NXS2/L, N = 4/group). The percentage of cells positive for CD4 or CD8 was determined by counting positive and total cells in 20 fields in each tumor section at 200× magnification using the formula: number of positive cells/total cell number × 100 (Mann–Whitney U test; *p < 0.05).

Download figure to PowerPoint

NB-derived soluble Gal-1 induces T cell apoptosis and inhibits DC maturation

Because Gal-1 has been reported to modulate T cell survival,26, 27 we analyzed the capacity of NB supernatants corresponding to transfectants secreting low or high Gal-1 levels, to induce apoptosis on splenic T cells. For this, we cultured MACS-sorted CD3+ T cells 24 hr in NXS2/L or NXS2/H SFCM and analyzed annexin V/PI staining as well as caspase 3 expression. We found a 28.15% increase in annexin V+ cells on T cells exposed to NXS2/H SFCM when compared with T cells preincubated with NXS2/L SFCM (Fig. 6a). Additionally, we found enhanced expression of active caspase 3 in T cells exposed to NXS2/H SFCM (Fig. 6b), suggesting that NB cells secrete Gal-1 to dampen effector T cell responses by inducing T cell apoptosis

thumbnail image

Figure 6. Effect of NB-derived Gal-1 on T cell apoptosis and on DC maturation. (a) The % of annexin V+ T cells was determined by flow cytometry. Annexin V/PI staining was performed on isolated CD3+ T cells. Data (mean ± S.E.M.) are shown from two independent experiments and considered statistically significant (unpaired t test; * p < 0.05, ** p ≤ 0.01). (b) Active caspase 3 was determined by Western blot analysis. Graph shows representative results from two independent experiments. GAPDH expression was used as control. (c) DC maturation was examined by analyzing the expression levels of I-A/I-K (murine MHC class II) and CD86 by flow cytometry. Bone marrow-derived DCs were incubated for 24 hr with SFCM from NXS2, NXS2/H and NXS2/L. DCs cultured with DC medium alone (DCM) were used as negative control. Upper panel shows data from three independent experiments (mean ± S.E.M.) (unpaired t test; *p < 0.05, **p ≤ 0.01). Lower panels illustrate density plots of one representative experiment.

Download figure to PowerPoint

Because Gal-1 not only inhibits T cell effector functions but also controls the DC compartment,18 we next assessed the effects of NB-derived Gal-1 on DC maturation as an additional inhibitory mechanisms. For this, we cultured bone marrow-derived DC from A/J mice in SFCM obtained from cultures of NXS2/L (Gal-1low) or NXS2/H, NXS2 (Gal-1high) cells. Remarkably, supernatants containing high amounts of Gal-1 (NXS2/H, NXS2) showed substantially lower expression of the activation/maturation markers I-A/I-K (25.8%) and CD86 (25.6%) on CD11c+ BMDCs when compared with CD11c+ BMDCs (57.3% and 50.8%) exposed to normal differentiation and maturation conditions (Fig. 6c). Of note, an increased frequency of CD11c+ BMDCs was positive for I-A/I-K (71%) and CD86 (35.8%) when cultured in the presence of SFCM from NXS2/L (Gal-1low) cells (Fig. 6c). Thus, NB-derived Gal-1 may control effector T cell responses during NB progression either directly via induction of T cell apoptosis or indirectly via promotion of immature or tolerogenic DCs.

Discussion

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The poor prognosis of children diagnosed with stage 4 NB demands a better understanding of NB escape mechanisms as a basis for the development of effective immunotherapeutic approaches. Our results identified a novel mechanism of tumor-immune escape in NB mediated by tumor-derived Gal-1, which favors immune cell dysfunction leading to tumor progression and metastasis. The regulatory function of Gal-1 seems to involve T cell apoptosis and the suppression of DC maturation which favor a tolerogenic microenvironment in NB.

Although not extensively studied like other tumor types, it has been proposed that NB suppresses activation and survival of effector T lymphocytes through expression of Fas L,28 downregulation of MHC class I expression29 and surface expression of inhibitory cell surface molecules, such as CTLA-4, B7-H3 and CD200.30–32 Furthermore, NB cells secrete inhibitory molecules targeting monocytes, NK and T cells that are involved in efficient immune cell control, i.e. the MHC class I chain-related gene A protein (sMICA), sHLA-G, the macrophage migration inhibitory factor (MIF) and TGF-β.33–36 Moreover, immune escape mechanisms developed by NB were additionally linked to unfavorable prognostic amplification of N-MYC, which was demonstrated to repress monocyte chemoattractant protein-1/CC chemokine ligand 2 (MCP-1/CCL2) and influence local NKT cell recruitment.37 Recently, the expression of the neurotropin receptor TrkB, also associated with aggressive NB, was found to correlate with intracellular levels of Gal-1 in primary NB.21 Here, we found expression of Gal-1 in additional human (LAN-1, Kelly, SK-N-AS and SK-N-SH) and murine (NXS2) NB cell lines and demonstrated that this soluble lectin is actively secreted by NB cells. To assess the role of Gal-1 in NB-driven immune escape mechanisms, we generated knockdown cells expressing low levels of Gal-1 (NXS2/L). Inoculation of Gal-1 knockdown transfectants into syngeneic immunocompetent A/J mice (H2-KkDd) led to reduced tumor growth and impaired metastatic potential when compared with mice receiving NB cells expressing high amounts of Gal-1 (NXS2, NXS2/H). These results demonstrate for the first time the important role of Gal-1 as a tolerogeneic molecule in NB and are in line with our previous findings on Gal-1-dependent tumor progression.12–14 Importantly, no significant differences in the viability and proliferation of NXS2/L (Gal-1low) and NXS2, NXS2/H (Gal-1high) cells could be observed in vitro, suggesting that fluctuations in the levels of Gal-1 do not affect intrinsic growth or survival of tumor cells. Supporting a role for Gal-1 in mediating NB-induced immunosuppression, T cells from NB tumors expressing low Gal-1 levels showed augmented cytotoxic activity when compared with T cells from mice harboring NB tumors expressing high levels of this lectin. This effect may reflect Gal-1 crosslinking of glycoproteins or glycolipids expressed on T cells including CD43, CD45, CD7 or GM1.38 Moreover, Gal-1 has been shown to block the production of proinflammatory Th1 and Th17 cytokines.22 In our model, T cells from NXS2/L-challenged mice secreted higher levels of IFN-γ than T cells from mice receiving NB tumors expressing high Gal-1 levels. IFN-γ is known to upregulate MHC class I and costimulatory molecules on NB,39 an effect which may unleash presentation of NB-associated antigens to CD4+ type-1 Th1 and CD8+ cytotoxic T cells.40 Interestingly, silencing Gal-1 expression resulted in enhanced T cell migration in vitro and increased CD4+ and CD8+ T cells in the tumor parenchyma, suggesting enhanced recruitment or survival of these cells in primary NXS2/L tumors compared to NXS2 or NXS2/H tumors. In this regard, blockade of Gal-1 has been proposed as adjuvant to classical immunotherapeutic modalities, and different galectin inhibitors have been designed and tested in vitro and in vivo.41–44 Combining Gal-1-specific disaccharides with classical vaccination protocols significantly decreased tumor progression and improved survival in a mice model of breast cancer.45 Hence, the efficiency of recently developed DNA vaccines targeting NB antigens, i.e. tyrosin hydroxylase (TH), disialoganglioside GD-2 or survivin24, 25, 46 might be improved by blocking Gal-1 expression.

Importantly, Gal-1 has been shown to induce growth arrest and apoptosis of activated T cells.26, 27 Here, we confirmed these findings as NB-derived Gal-1 was able to increase phosphatidylserine exposure and induce expression of active caspase 3 on T cells. To further explore the mechanisms underlying the tolerogenic effects of Gal-1 in NB, we next studied the influence of tumor-derived Gal-1 on DC maturation. Immature or tolerogenic DCs actively induce T cell unresponsiveness through different mechanisms involving induction of T cell anergy, secretion of anti-inflammatory cytokines (like IL-10) or induction of FoxP3+ or FoxP3- T regulatory (Treg) cells which in turn block the expansion or activation of antigen-specific T cells.47, 48 Here, supernatants from NXS2 NB cells induced the differentiation of DCs bearing an immature phenotype characterized by reduced surface expression of MHC class I (I-A/I-K) and CD86. Blocking Gal-1 expression on NB cells prevented this effect as supernatant from NXS2/L allowed full maturation of DCs in culture. This result shows that Gal-1 has a direct influence on the maturation state of DCs, resulting in impaired antigen presentation and effector immune responses, as has been demonstrated in other tumor types.49 Immature DCs usually lead to the differentiation and/or expansion of Treg cells which support tumor dissemination. In fact, Treg cell depletion in vivo effectively improved immune responses against NB.50 Notably, a functional cross-talk between Gal-1, IL-27-producing DCs and IL-10 producing Treg cells has recently been demonstrated in different tumor and autoimmune settings.18

In summary, we provide here the first evidence demonstrating a role for Gal-1 as a pivotal regulator of immune response in NB. We found that cancer-derived Gal-1 modulates T cell and DC compartments, resulting in increased tumor growth and metastasis. Thus, Gal-1 represents a novel target, either alone or in combination with other therapeutic modalities, for the treatment of one of the most challenging tumors during childhood.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was mainly supported by grants from the German-Reseach Foundation, DFG (Fe 914/2-1) to SF and the Walter-Schulz-Foundation (SF and ACZ). SF further received a price from HEXAL (Barleben, Sachsen-Anhalt). RS received a grant from the Kind-Philipp-Foundation and EB was supported by a training grant from the Medical Faculty, Otto-von-Guericke University. This work was distinguished with the AACR Scholar-in-Training Award for RS. GAR was funded by the Agencia Nacional de Promoción Científica y Tecnológica (FONCYT; Argentina), the University of Buenos Aires, Fundación Sales (Argentina), Prostate Cancer Research Foundation (UK) and Mizutani Foundation for Glycoscience (Japan). MS was supported by the Argentinean National Research Council (CONICET). Further support was provided by the NGFNplus program within the ENGINE consortium to HNL (01GS0899).

References

  1. Top of page
  2. Abstract
  3. Material and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Maris JM, Hogarty MD, Bagatell R, Cohn SL. Neuroblastoma. Lancet 2007; 369: 210620.
  • 2
    Cohn SL, Pearson AD, London WB, Monclair T, Ambros PF, Brodeur GM, Faldum A, Hero B, Iehara T, Machin D, Mosseri V, Simon T, et al. The International Neuroblastoma Risk Group (INRG) classification system: an INRG Task Force report. J Clin Oncol 2009; 27: 28997.
  • 3
    Berthold F, Boos J, Burdach S, Erttmann R, Henze G, Hermann J, Klingebiel T, Kremens B, Schilling FH, Schrappe M, Simon T, Hero B. Myeloablative megatherapy with autologous stem-cell rescue versus oral maintenance chemotherapy as consolidation treatment in patients with high-risk neuroblastoma: a randomised controlled trial. Lancet Oncol 2005; 6: 64958.
  • 4
    Pearson AD, Pinkerton CR, Lewis IJ, Imeson J, Ellershaw C, Machin D. High-dose rapid and standard induction chemotherapy for patients aged over 1 year with stage 4 neuroblastoma: a randomised trial. Lancet Oncol 2008; 9: 24756.
  • 5
    Gray JC, Kohler JA. Immunotherapy for neuroblastoma: turning promise into reality. Pediatr Blood Cancer 2009; 53: 93140.
  • 6
    Yu AL, Gilman AL, Ozkaynak MF, London WB, Kreissman SG, Chen HX, Smith M, Anderson B, Villablanca JG, Matthay KK, Shimada H, Grupp SA, et al. Anti-GD2 antibody with GM-CSF, interleukin-2, and isotretinoin for neuroblastoma. N Engl J Med 2010; 363: 132434.
  • 7
    Rabinovich GA, Gabrilovich D, Sotomayor EM. Immunosuppressive strategies that are mediated by tumor cells. Annu Rev Immunol 2007; 25: 26796.
  • 8
    Camby I, Le MM, Lefranc F, Kiss R. Galectin-1: a small protein with major functions. Glycobiology 2006; 16: 137R57R.
  • 9
    Le Mercier M, Fortin S, Mathieu V, Roland I, Spiegl-Kreinecker S, Haibe-Kains B, Bontempi G, Decaestecker C, Berger W, Lefranc F, Kiss R. Galectin 1 proangiogenic and promigratory effects in the Hs683 oligodendroglioma model are partly mediated through the control of BEX2 expression. Neoplasia 2009; 11: 48596.
  • 10
    Liu FT, Rabinovich GA. Galectins as modulators of tumour progression. Nat Rev Cancer 2005; 5: 2941.
  • 11
    Liu FT, Rabinovich GA. Galectins: regulators of acute and chronic inflammation. Ann N Y Acad Sci 2010; 1183: 15882.
  • 12
    Rubinstein N, Alvarez M, Zwirner NW, Toscano MA, Ilarregui JM, Bravo A, Mordoh J, Fainboim L, Podhajcer OL, Rabinovich GA. Targeted inhibition of galectin-1 gene expression in tumor cells results in heightened T cell-mediated rejection; A potential mechanism of tumor-immune privilege. Cancer Cell 2004; 5: 24151.
  • 13
    Juszczynski P, Ouyang J, Monti S, Rodig SJ, Takeyama K, Abramson J, Chen W, Kutok JL, Rabinovich GA, Shipp MA. The AP1-dependent secretion of galectin-1 by Reed Sternberg cells fosters immune privilege in classical Hodgkin lymphoma. Proc Natl Acad Sci USA 2007; 104: 131349.
  • 14
    Kuo PL, Hung JY, Huang SK, Chou SH, Cheng DE, Jong YJ, Hung CH, Yang CJ, Tsai YM, Hsu YL, Huang MS. Lung cancer-derived galectin-1 mediates dendritic cell anergy through inhibitor of DNA binding 3/IL-10 signaling pathway. J Immunol 2011; 186: 152130.
  • 15
    Banh A, Zhang J, Cao H, Bouley DM, Kwok S, Kong C, Giaccia AJ, Koong AC, Le QT. Tumor galectin-1 mediates tumor growth and metastasis through regulation of T-cell apoptosis. Cancer Res 2011; 71: 442331.
  • 16
    Tang D, Yuan Z, Xue X, Lu Z, Zhang Y, Wang H, Chen M, An Y, Wei J, Zhu Y, Miao Y, Jiang K. High expression of Galectin-1 in pancreatic stellate cells plays a role in the development and maintenance of an immunosuppressive microenvironment in pancreatic cancer. Int J Cancer 2011. [DOI: 10.1002/ijc.26290].
  • 17
    Toscano MA, Bianco GA, Ilarregui JM, Croci DO, Correale J, Hernandez JD, Zwirner NW, Poirier F, Riley EM, Baum LG, Rabinovich GA. Differential glycosylation of TH1, TH2 and TH-17 effector cells selectively regulates susceptibility to cell death. Nat Immunol 2007; 8: 82534.
  • 18
    Ilarregui JM, Croci DO, Bianco GA, Toscano MA, Salatino M, Vermeulen ME, Geffner JR, Rabinovich GA. Tolerogenic signals delivered by dendritic cells to T cells through a galectin-1-driven immunoregulatory circuit involving interleukin 27 and interleukin 10. Nat Immunol 2009; 10: 98191.
  • 19
    Camby I, Belot N, Lefranc F, Sadeghi N, de LY, Kaltner H, Musette S, Darro F, Danguy A, Salmon I, Gabius HJ, Kiss R. Galectin-1 modulates human glioblastoma cell migration into the brain through modifications to the actin cytoskeleton and levels of expression of small GTPases. J Neuropathol Exp Neurol 2002; 61: 58596.
  • 20
    Le Mercier M, Fortin S, Mathieu V, Kiss R, Lefranc F. Galectins and gliomas. Brain Pathol 2010; 20: 1727.
  • 21
    Cimmino F, Schulte JH, Zollo M, Koster J, Versteeg R, Iolascon A, Eggert A, Schramm A. Galectin-1 is a major effector of TrkB-mediated neuroblastoma aggressiveness. Oncogene 2009; 28: 201523.
  • 22
    Rabinovich GA, Daly G, Dreja H, Tailor H, Riera CM, Hirabayashi J, Chernajovsky Y. Recombinant galectin-1 and its genetic delivery suppress collagen-induced arthritis via T cell apoptosis. J Exp Med 1999; 190: 38598.
  • 23
    Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N, Schuler G. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 1999; 223: 7792.
  • 24
    Fest S, Huebener N, Weixler S, Bleeke M, Zeng Y, Strandsby A, Volkmer-Engert R, Landgraf C, Gaedicke G, Riemer AB, Michalsky E, Jaeger IS, et al. Characterization of GD2 peptide mimotope DNA vaccines effective against spontaneous neuroblastoma metastases. Cancer Res 2006; 66: 1056775.
  • 25
    Fest S, Huebener N, Bleeke M, Durmus T, Stermann A, Woehler A, Baykan B, Zenclussen AC, Michalsky E, Jaeger IS, Preissner R, Hohn O, et al. Survivin minigene DNA vaccination is effective against neuroblastoma. Int J Cancer 2009; 125: 10414.
  • 26
    Perillo NL, Pace KE, Seilhamer JJ, Baum LG. Apoptosis of T cells mediated by galectin-1. Nature 1995; 378: 7369.
  • 27
    Blaser C, Kaufmann M, Muller C, Zimmermann C, Wells V, Mallucci L, Pircher H. Beta-galactoside-binding protein secreted by activated T cells inhibits antigen-induced proliferation of T cells. Eur J Immunol 1998; 28: 23119.
  • 28
    Shurin GV, Gerein V, Lotze MT, Barksdale EM, Jr. Apoptosis induced in T cells by human neuroblastoma cells: role of Fas ligand. Nat Immun 1998; 16: 26374.
  • 29
    Wolfl M, Jungbluth AA, Garrido F, Cabrera T, Meyen-Southard S, Spitz R, Ernestus K, Berthold F. Expression of MHC class I, MHC class II, and cancer germline antigens in neuroblastoma. Cancer Immunol Immunother 2005; 54: 4006.
  • 30
    Xu H, Cheung IY, Guo HF, Cheung NK. MicroRNA miR-29 modulates expression of immunoinhibitory molecule B7-H3: potential implications for immune based therapy of human solid tumors. Cancer Res 2009; 69: 627581.
  • 31
    Siva A, Xin H, Qin F, Oltean D, Bowdish KS, Kretz-Rommel A. Immune modulation by melanoma and ovarian tumor cells through expression of the immunosuppressive molecule CD200. Cancer Immunol Immunother 2008; 57: 98796.
  • 32
    Contardi E, Palmisano GL, Tazzari PL, Martelli AM, Fala F, Fabbi M, Kato T, Lucarelli E, Donati D, Polito L, Bolognesi A, Ricci F, et al. CTLA-4 is constitutively expressed on tumor cells and can trigger apoptosis upon ligand interaction. Int J Cancer 2005; 117: 53850.
  • 33
    Morandi F, Levreri I, Bocca P, Galleni B, Raffaghello L, Ferrone S, Prigione I, Pistoia V. Human neuroblastoma cells trigger an immunosuppressive program in monocytes by stimulating soluble HLA-G release. Cancer Res 2007; 67: 643341.
  • 34
    Yan X, Orentas RJ, Johnson BD. Tumor-derived macrophage migration inhibitory factor (MIF) inhibits T lymphocyte activation. Cytokine 2006; 33: 18898.
  • 35
    Raffaghello L, Prigione I, Airoldi I, Camoriano M, Levreri I, Gambini C, Pende D, Steinle A, Ferrone S, Pistoia V. Downregulation and/or release of NKG2D ligands as immune evasion strategy of human neuroblastoma. Neoplasia 2004; 6: 55868.
  • 36
    Fontana A, Bodmer S, Frei K, Malipiero U, Siepl C. Expression of TGF-beta 2 in human glioblastoma: a role in resistance to immune rejection? Ciba Found Symp 1991; 157: 2328.
  • 37
    Song L, Ara T, Wu HW, Woo CW, Reynolds CP, Seeger RC, DeClerck YA, Thiele CJ, Sposto R, Metelitsa LS. Oncogene MYCN regulates localization of NKT cells to the site of disease in neuroblastoma. J Clin Invest 2007; 117: 270212.
  • 38
    Rabinovich GA, Toscano MA. Turning ‘sweet’ on immunity: galectin-glycan interactions in immune tolerance and inflammation. Nat Rev Immunol 2009; 9: 33852.
  • 39
    Sigal RK, Lieberman MD, Reynolds JV, Shou J, Ziegler MM, Daly JM. Low-dose interferon gamma renders neuroblastoma more susceptible to interleukin-2 immunotherapy. J Pediatr Surg 1991; 26: 38995.
  • 40
    Muranski P, Restifo NP. Adoptive immunotherapy of cancer using CD4(+) T cells. Curr Opin Immunol 2009; 21: 2008.
  • 41
    Giguere D, Bonin MA, Cloutier P, Patnam R, St-Pierre C, Sato S, Roy R. Synthesis of stable and selective inhibitors of human galectins-1 and −3. Bioorg Med Chem 2008; 16: 781123.
  • 42
    Ingrassia L, Camby I, Lefranc F, Mathieu V, Nshimyumukiza P, Darro F, Kiss R. Anti-galectin compounds as potential anti-cancer drugs. Curr Med Chem 2006; 13: 351327.
  • 43
    Rabinovich GA, Cumashi A, Bianco GA, Ciavardelli D, Iurisci I, D'Egidio M, Piccolo E, Tinari N, Nifantiev N, Iacobelli S. Synthetic lactulose amines: novel class of anticancer agents that induce tumor-cell apoptosis and inhibit galectin-mediated homotypic cell aggregation and endothelial cell morphogenesis. Glycobiology 2006; 16: 21020.
  • 44
    Salomonsson E, Thijssen VL, Griffioen AW, Nilsson UJ, Leffler H. The anti-angiogenic peptide anginex greatly enhances galectin-1 binding affinity for glycoproteins. J Biol Chem 2011; 286: 138014.
  • 45
    Stannard KA, Collins PM, Ito K, Sullivan EM, Scott SA, Gabutero E, Darren G, I, Low P, Nilsson UJ, Leffler H, Blanchard H, Ralph SJ. Galectin inhibitory disaccharides promote tumour immunity in a breast cancer model. Cancer Lett 2010; 299: 95110.
  • 46
    Huebener N, Fest S, Strandsby A, Michalsky E, Preissner R, Zeng Y, Gaedicke G, Lode HN. A rationally designed tyrosine hydroxylase DNA vaccine induces specific antineuroblastoma immunity. Mol Cancer Ther 2008; 7: 224151.
  • 47
    Li L, Li SP, Min J, Zheng L. Hepatoma cells inhibit the differentiation and maturation of dendritic cells and increase the production of regulatory T cells. Immunol Lett 2007; 114: 3845.
  • 48
    Martin P, Del Hoyo GM, Anjuere F, Arias CF, Vargas HH, Fernandez L, Parrillas V, Ardavin C. Characterization of a new subpopulation of mouse CD8alpha+ B220+ dendritic cells endowed with type 1 interferon production capacity and tolerogenic potential. Blood 2002; 100: 38390.
  • 49
    Walker SR, Redlinger RE, Jr., Barksdale EM, Jr. Neuroblastoma-induced inhibition of dendritic cell IL-12 production via abrogation of CD40 expression. J Pediatr Surg 2005; 40: 2449.
  • 50
    Johnson BD, Jing W, Orentas RJ. CD25+ regulatory T cell inhibition enhances vaccine-induced immunity to neuroblastoma. J Immunother 2007; 30: 20314.