Tumor mRNA-transfected dendritic cells stimulate the generation of CTL that recognize neuroblastoma-associated antigens and kill tumor cells: immunotherapeutic implications

Abstract

Several observations suggest a potential role of T-cell-mediated immunity in the control of neuroblastoma (NB). However, the generation of NB-specific cytotoxic T lymphocytes (CTL) on T-cell priming with tumor mRNA-transfected dendritic cells (DC) has never been investigated before. In the present study, the feasibility of this strategy has been analyzed, both in healthy donors and in NB patients. Monocyte-derived DC were raised from three human leukocyte antigen (HLA) A2+ NB patients and seven HLA-A1+ or HLA-A2+ healthy donors transfected with mRNA from four NB cell lines and cocultured with autologous CD8+ lymphocytes. Expanded CTL expressed an effector/memory phenotype and a T cytotoxic 1-like profile of cytokine secretion. CTL specificity was demonstrated by interferon-gamma release on incubation with HLA-matched NB cell lines. The latter cell lines, but not autologous T-cell blasts, were lysed by CTL in an HLA-restricted manner. Cytotoxicity was found to involve the release of granzyme B. When tested for reactivity against NB-associated antigens, CTL from normal individuals recognized anaplastic lymphoma-associated kinase (ALK) and preferentially expressed antigen of melanoma (PRAME) peptides only, whereas patients' CTL reacted also to survivin, telomerase, and tyrosine hydroxylase peptides. This study demonstrates that DC transfected with NB mRNA induce the generation of patients' CTL specific for different NB-associated antigens, supporting the feasibility of NB T-cell immunotherapy.

Figure 1

Characterization of mature DC transfected with mRNA from NB cell lines. (A) Mature DC derived from NB patients and normal donors were transfected with pooled mRNA extracted from four NB cell lines and analyzed for the presence of tumor-derived TH transcripts by qualitative and real-time RT-PCR. Left panel: RT-PCR analysis of TH gene expression was performed on total RNA extracted from DC before and after transfection (1 and 24 hours); pooled mRNA from NB cell lines was used as positive control. CTR-negative control; MW, molecular weight markers. Right panel: Real-time RT-PCR analysis of TH gene expression was performed on total RNA extracted from DC before and 1 hour after transfection. Results are expressed as nanomoles of TH per picogram of RNA. (B) Mature DC were divided into two aliquots, the first of which was transfected with NB cell line mRNA and cultured for 24 hours, whereas the second unmanipulated aliquot was kept in culture for the same time. The immunophenotype of these cell fractions was next investigated by flow cytometry. The ranges of the results obtained with all patients tested are shown. Results are expressed as percentages of positive cells. The histograms of a representative experiment performed with cells from patient 2 are shown in (C). Empty profiles indicate staining with isotype-matched control immunoglobulins; black profiles refer to staining with specific mAb.

Figure 2

Immunophenotype of in vitro-expanded CD8⁺ T lymphocytes upon priming with autologous DC transfected with NB mRNA. The immunophenotypic profile of in vitro-expanded CTL was assessed by flow cytometry using fluorochrome-conjugate mAbs. Intracellular cytokine staining was performed as detailed in Gattorno et al. [37]. (A) The ranges of the results from all patients tested are shown. Results are expressed as percentages of positive cells. (B) The histograms of a representative experiment performed with cells from patient 2 are shown. Empty profiles indicate staining with isotype-matched control immunoglobulins; black profiles refer to staining with specific mAbs. Numbers on the top right refer to percentages of positive cells.

Figure 3

NB specificity of CTL generated on priming with autologous DC transfected with NB mRNA. CD8⁺ T-cell lines derived from two NB patients (patients 1 and 2, both HLA-A2⁺) and five healthy donors (donors 3, 5, and 6, HLA-A2⁺; donors 2 and 7, HLA-A1⁺) were assessed for IFN-γ release by ELISPOT assays. T cells were cultured with medium alone (medium) or with HLA-matched NB cell lines at a 1:2 cell ratio, in the absence (-) or in the presence (+) of anti-HLA class I mAb. Indicated spot numbers per seeded lymphocyte represent the mean values of three replicates. One of three representative experiments performed for each subject is shown.

Figure 4

NB cell lysis by CTL expanded in vitro using autologous DC transfected with NB mRNA. CTL expanded from two NB patients (patients 1 and 2, both HLA-A2⁺) and five healthy donors (donors 1, 3, 4, and 5, HLA-A2⁺; donor 7, HLA-A1⁺) were tested for cytotoxic activity in 4-hour ⁵¹Cr release assays against HLA-matched NB cell lines or autologous T-cell blasts in the absence (-) or in the presence (+) of anti-HLA class I mAb. Results are expressed as percent-specific lysis at a 50:1 effector/target ratio. Numbers represent the mean values of three replicates. One of three representative experiments performed for each subject is shown.

Figure 5

Granzyme B release by NB-specific CTL on incubation with HLA-matched NB cell lines. CTL expanded from two NB patients (patients 1 and 2, both HLA-A2⁺) and two healthy donors (donors 4 and 5, HLA-A2⁺) were tested for granzyme B release by ELISPOT assays. T cells were cultured with medium alone (medium) or with HLA-matched NB cell lines at a 1:2 cell ratio. Indicated spot numbers per seeded lymphocyte represent the mean values of three replicates. One of three representative experiments performed for each subject is shown.

Figure 6

NB-specific CTL recognize multiple TAA. CTL from three HLA-A2⁺ NB patients (patients 1, 2, and 3) and three HLA-A2⁺ healthy donors (donors 4, 5, and 6) were assayed by IFN-γ ELISPOT using as target a T2 cell line pulsed with HLA-A2 synthetic peptides of survivin, PRAME, ALK, hTERT, and TH, respectively. Indicated spot numbers per seeded lymphocyte represent the mean values of three replicates after the subtraction of spots obtained with T2 cells pulsed with an irrelevant control peptide (GFP). One of three representative experiments performed for each subject is shown.