Ewing sarcoma dissemination and response to T-cell therapy in mice assessed by whole-body magnetic resonance imaging.

BACKGROUND: Novel treatment strategies in Ewing sarcoma include targeted cellular therapies. Preclinical in vivo models are needed that reflect their activity against systemic (micro)metastatic disease. METHODS: Whole-body magnetic resonance imaging (WB-MRI) was used to monitor the engraftment and dissemination of human Ewing sarcoma xenografts in mice. In this model, we evaluated the therapeutic efficacy of T cells redirected against the Ewing sarcoma-associated antigen GD2 by chimeric receptor engineering. RESULTS: Of 18 mice receiving intravenous injections of VH-64 Ewing sarcoma cells, all developed disseminated tumour growth detectable by WB-MRI. All mice had lung tumours, and the majority had additional manifestations in the bone, soft tissues, and/or kidney. Sequential scans revealed in vivo growth of tumours. Diffusion-weighted whole-body imaging with background signal suppression effectively visualised Ewing sarcoma growth in extrapulmonary sites. Animals receiving GD2-targeted T-cell therapy had lower numbers of pulmonary tumours than controls, and the median volume of soft tissue tumours at first detection was lower, with a tumour growth delay over time. CONCLUSION: Magnetic resonance imaging reliably visualises disseminated Ewing sarcoma growth in mice. GD2-retargeted T cells can noticeably delay tumour growth and reduce pulmonary Ewing sarcoma manifestations in this aggressive disease model.

The ganglioside antigen G(D2) is surface-expressed in Ewing sarcoma and allows for MHC-independent immune targeting.

BACKGROUND: Novel treatment strategies are needed to cure disseminated Ewing sarcoma. Primitive neuroectodermal features and a mesenchymal stem cell origin are both compatible with aberrant expression of the ganglioside antigen G(D2) and led us to explore G(D2) immune targeting in this cancer. METHODS: We investigated G(D2) expression in Ewing sarcoma by immunofluorescence staining. We then assessed the antitumour activity of T cells expressing a chimeric antigen receptor specific for G(D2) against Ewing sarcoma in vitro and in vivo. RESULTS: Surface G(D2) was detected in 10 out of 10 Ewing sarcoma cell lines and 3 out of 3 primary cell cultures. Moreover, diagnostic biopsies from 12 of 14 patients had uniform G(D2) expression. T cells specifically modified to express the G(D2)-specific chimeric receptor 14. G2a-28ζ efficiently interacted with Ewing sarcoma cells, resulting in antigen-specific secretion of cytokines. Moreover, chimeric receptor gene-modified T cells from healthy donors and from a patient exerted potent, G(D2)-specific cytolytic responses to allogeneic and autologous Ewing sarcoma, including tumour cells grown as multicellular, anchorage-independent spheres. G(D2)-specific T cells further had activity against Ewing sarcoma xenografts. CONCLUSION: G(D2) surface expression is a characteristic of Ewing sarcomas and provides a suitable target antigen for immunotherapeutic strategies to eradicate micrometastatic cells and prevent relapse in high-risk disease.

Lack of expression of the chondroitin sulphate proteoglycan neuron-glial antigen 2 on candidate stem cell populations in paediatric acute myeloid leukaemia/abn(11q23) and acute lymphoblastic leukaemia/t(4;11).

It has increasingly been acknowledged that only a few leukaemic cells possess the capability to renew themselves and that only these self-renewing leukaemic stem cells are able to initiate relapses. Therefore, these leukaemic stem cells should be the target cells for therapy and for minimal residual disease (MRD) detection. Because of its presence on blasts of 11q23-rearranged high-risk leukaemic patients, neuron-glial antigen 2 (NG2) is thought to be a valuable marker for detecting leukaemic stem cells. Six acute myeloid leukaemia (AML)/abn(11q23) and three acute lymphoblastic leukaemia (ALL)/t(4;11) samples were analysed by four-colour flow cytometry for NG2 expression on primitive cell populations. Candidate leukaemic cell populations were defined by the antigen profiles CD34+CD38- in AML and CD34+CD19-CD117+ in ALL. Surprisingly, in all patients these candidate stem cell populations were shown to lack expression of NG2. Instead, a correlation between the expression of the myeloid differentiation marker CD33 and increasing levels of NG2 on maturing cells could be demonstrated. Similarly, in ALL patients CD34+CD19+ cells showed a higher expression of NG2 mRNA compared with CD34+CD19-. Thus, NG2 appears to be upregulated with differentiation and not to be expressed on primitive disease-maintaining cells. This hampers the clinical use of NG2 as a therapeutic target and as a sensitive marker for MRD detection.

PI3K/AKT is involved in mediating survival signals that rescue Ewing tumour cells from fibroblast growth factor 2-induced cell death.

While in vitro studies had shown that fibroblast growth factor 2 (FGF2) can induce cell death in Ewing tumours, it remained unclear how Ewing tumour cells survive in vivo within a FGF2-rich microenvironment. Serum- and integrin-mediated survival signals were, therefore, studied in adherent monolayer and anchorage-independent colony cell cultures. In a panel of Ewing tumour cell lines, either adhesion to collagen or exposure to serum alone only had a minor protective effect against FGF2. However, both combined led to complete resistance to 5 ng ml(-1) FGF2 in three of four FGF2-sensitive cell lines (RD-ES, RM-82 and WE-68), and to an increased survival as compared to other culture conditions in TC-71 cells. Inhibition studies with LY294002 demonstrated that the serum signal is mediated via the phosphoinositide 3-OH kinase/AKT pathway. Thus, Ewing tumour cells escape FGF2-induced cell death by modulating FGF2 signalling. The tumour microenvironment provides the necessary survival signals by integrin-mediated adhesion and soluble serum factor(s). These survival signals warrant further investigation as a potential resistance mechanism to other apoptosis-inducing agents in vivo.

Establishment of an in vivo model for pediatric Ewing tumors by transplantation into NOD/scid mice.

Ewing tumors are a clinically heterogeneous group of childhood sarcomas that represent a paradigm for understanding solid tumor biology, as they are the first group of sarcomas for which a chromosome translocation has been characterized at the molecular level. However, the biologic organization of the tumor, especially the processes that govern proliferation, differentiation, and metastasis of primitive tumor stem cells is poorly understood. Therefore, to develop a biologically relevant in vivo model, five different Ewing tumor cell lines and primary tumor cells from three patients were transplanted into immune-deficient mice via intravenous injection. NOD/scid mice that carry a complex immune deficiency and thus nearly completely lack the ability to reject human cells were used as recipients. Overall, 26 of 52 mice (50%) transplanted with VH-64, WE-68, CADO-ES1, TC-71, and RM-82 cells and 4 of 10 mice (40%) transplanted with primary tumor cells engrafted. Moreover, primary cells that did not grow in vitro proliferated in mice. The pattern of metastasis was similar to that in patients with frequent metastases in lungs (62%), bone marrow (30%), and bone (23%). Using limiting dilution experiments, the frequency of the engraftment unit was estimated at 1 Ewing tumor-initiating cell in 3 x 10(5) VH-64 cells. These data demonstrate that we have been able to establish an in vivo model that recapitulates many aspects of growth and progression of human Ewing tumors. For the first time, this model provides the opportunity to identify and characterize primitive in vivo clonogenic solid tumor stem cells. This model will, therefore, be instrumental in studying many aspects of tumor cell biology, including organ-selective metastasis and tumor angiogenesis.

Interferon-gamma increases IL-6 production in human glioblastoma cell lines.

BACKGROUND: Various immunotherapeutical approaches are presently evaluated for their efficacy to eradicate glioma cells. Complicating the concepts, these tumors secrete cytokines which modulate the immune response. MATERIALS & METHODS: We analyzed the influence of interferon gamma (IFN-gamma) on the cytokine production and IFN-gamma receptor expression in T98G and U87-MG human glioma cells. RESULTS: The IFN-gamma receptors were own-regulated after IFN-gamma treatment. Secretion of interleukin-6 (IL-6) protein was elevated by factors of 6.4 in T98G cells and 5.2 in U87-MG. Other cytokines were increased as well, but less constantly: IL-8, and VEGF were elevated significantly only in T98G, but not in U87-MG. Increases of IL-1 beta, IL-1 alpha and TGF beta-2 were only detectable at the mRNA level. TNF was not detectable in any of the cell lines, and TGF-beta, alpha FGF and HG were not influenced by IFN-gamma. CONCLUSION: IL-6 produced by glioma cells in response to IFN-gamma might support immune eradication of glioma cells.

Def-2, -3, -6 and -8, novel mouse genes differentially expressed in the haemopoietic system.

To identify developmentally regulated genes during myeloid differentiation, a self-inactivating retroviral gene-trap vector carrying a beta-galactosidase-neomycin (SA/lacZ/neo) fusion gene was constructed and used to infect myeloid progenitor cells (FDCP-Mix A4). G418-resistant and beta-galactosidase positive cell lines (gene-trap integration [GTI] clones) were established and induced to differentiate in vitro into either macrophages or granulocytes. Expression of the trapped loci was monitored at a single-cell level by analysing the mature cell types for beta-galactosidase activity. All 37 GTI clones tested showed down-regulation either during granulocyte or both granulocytic and macrophage differentiation. The endogenous coding regions fused to the SA/lacZ/neo reporter gene were isolated from eight clones. Molecular analysis revealed that half of them represented novel mouse genes (def-2, -3, -6 and -8) which we confirmed to be differentially expressed in primary haemopoietic tissues. Database searches revealed no significant similarities for def-2 (associated with haemopoietic progenitors) and def-8 (expressed most strongly in peripheral leucocytes). Def-6, which is down-regulated upon the differentiation into myeloid as well as erythroid lineages, was found to be closely related but not identical with the recently described B-cell-specific switch recombinase SWAP-70. Def-3, which is down-regulated upon differentiation into granulocytes but expressed in progenitor cells and macrophages, defines a novel family of RNA binding proteins.

CD44 expression and hyaluronic acid binding of malignant glioma cells.

The mechanisms leading to rapid invasive growth of malignant gliomas are poorly understood. Expression of the hyaluronic acid (HA) receptor CD44 and adhesion to HA are involved in invasive properties. Our previous studies have shown that malignant glioma cells are able to adhere to extracellular HA. Here we investigated expression of the hyaluronic acid receptor CD44 protein in five human (T98G, A172, U87MG, 86HG39, 85HG66) and two rat (C6, 9L) glioma cell lines. Influence of anti-CD44 antibody and hyaluronidase-preincubation on the HA-binding was determined using HA/BSA (bovine serum albumin)-coated culture plates. While all gliomas were highly positive for CD44 with no differences in the number of positive staining cells, median fluorescence intensity decreased as follows: C6>T98G>9L>85HG66> 86HG39>A172>U87MG. Using HA/BSA coated culture plates the relative levels of specific adhesion to HA were determined as T98G>A172>9L>86HG39>U87MG> 85HG66. C6 cells failed to bind HA specifically. Incubation with anti-human-CD44 MAb significantly decreased HA-adhesion of T98G, A172, 85HG66 and U87MG human glioma cells. However the binding capacity was completely blocked only in 85HG66 cells. The three other cell lines kept a specific HA-adhesion after saturation of the receptor. Hyaluronidase pretreatment markedly enhanced HA-adhesion of C6 and 9L rat glioma cells. These results suggest that (i) HA-adhesion of malignant glioma cells is mainly, but not only, mediated by CD44, (ii) expression of CD44 does not correspond with adhesion capacity and (iii) cell-bound glycosaminoglycans may influence glioma cell adhesion to extracellular HA.

DEF-3(g16/NY-LU-12), an RNA binding protein from the 3p21.3 homozygous deletion region in SCLC.

DEF-3(g16/NY-LU-12) encodes a novel RNA binding protein isolated by positional cloning from an SCLC homozygous deletion region in 3p21.3 and, in parallel, as a differentially expressed gene during myelopoiesis from FDCPmix-A4 cells. DEF-3(g16/NY-LU-12) is ubiquitously expressed during mouse embryogenesis and in adult organs while human hematopoietic tissues showed differential expression. The mouse and human proteins are highly conserved containing two RNA recognition motifs (RRMs) and other domains associated with RNA binding and protein-protein interactions. A database search identified related proteins in human, rat, C. elegans and S. pombe including the 3p21.3 co-deleted gene, LUCA15. Recombinant proteins containing the RRMs of DEF-3(g16/NY-LU-12) and LUCA15 specifically bound poly(G) RNA homopolymers in vitro. These RRMs also show similarity to those of the Hu protein family. Since anti-Hu RRM domain antibodies are associated with an anti-tumor effect and paraneoplastic encephalomyelitis, we tested sera from Hu syndrome patients with the RRMs of DEF-3(g16/NY-LU-12) and LUCA15. These were non-reactive. Thus, DEF-3(g16/NY-LU-12) and LUCA15 represent members of a novel family of RNA binding proteins with similar expression patterns and in vitro RNA binding characteristics. They are co-deleted in some lung cancers and immunologically distinct from the Hu proteins.

P450-expression in brain tumors.

Oxazaphosphorines are inactive anticancer prodrugs that are bioactivated by hepatic cytochrome P450. Besides hepatic metabolism, there is increasing interest in the possibility of intratumoral activation of oxazaphosphorines by P450. Therefore, we investigated the expression of P450 (CYP3A4, CYP3A5, CYP2C9) by RT-PCR in 10 different brain tumor samples. Because P450 may be downregulated by interleukin-1 (IL-1) and IL-6, the receptors for IL-1 and IL-6 were analyzed. None of the brain tumors was positive for CYP3A4 whereas CYP3A5 was detected in 3 out of 10 tumors (two meningeomas, one medulloblastoma grade IV). All five gliomas, an ependymoma, and a lymphoma-metastase gave no signal. CYP2C9 mRNA was present in every sample studied. All samples were positive for IL-1 and IL-6 receptors. In summary, we have demonstrated that tumors of the CNS express P450, indicating that activation of prodrugs like oxazaphosphorines may take place intratumorally. However, the most abundantly hepatically expressed CYP3A4 enzyme is absent in the brain tumor samples. The presence of the IL-1 and IL-6 receptors opens the possibility that the wellknown downregulating influence of these cytokines also takes place in brain tumors.

Hyaluronic acid binding capacity of malignant glioma cells.

The mechanisms underlying the rapid invasive growth of malignant gliomas are poorly understood. Adhesion to extracellular hyaluronic acid (HA) has been implicated in the invasive properties of tumor cells. We investigated the HA binding capacity of human (T98G, A172, U87MG, 86HG39, 85HG66) and rat (C6, 9L) glioma cell lines by means of HA coated, bovine serum albumin (BSA)-blocked (HA/BSA) and only BSA-blocked culture plates. Results were compared with adhesion to native wells (100% adhesion). Adhesion to HA/BSA was high for T98G (84.4%), medium for 86HG39 (36%), 9L (33.1%), A172 (35.5%) and low for 85HG66 (21.3%) and U87MG (26.8%). Adhesion to only BSA-coated wells was significantly lower in all these cell lines, suggesting a specific HA-adhesion. Only C6 showed similar adhesion to HA/BSA and BSA alone, therefore, C6 failed to bind HA specifically. These results suggest that adhesion to extracellular HA might be involved in the invasion of some gliomas.

Diastereoselective synthesis of octahydro-14H-benzo[g]quinolino-[2,3-a]quinolidines. Improved cytotoxic activity against human brain tumor cell lines as a result of the increased rigidity of the molecular backbone.

Cis-Octahydro-14H-benzo[g]quinolino[2,3-a]quinolidines 6 were obtained in 6 steps from L-phenylalanine. The key step utilizes a diastereoselective intramolecular EtAlCl2-catalyzed hetero-Diels-Alder reaction. Compounds 6a-f were tested in vitro against human medulloblastoma D283 Med and glioblastoma A-172 and T98G cell lines and showed improved cytotoxicity compared to the corresponding, less rigid pyrido[1,2-b]isoquinolines 1.

Tumorangiogenesis: a network of cytokines.

During angiogenesis new blood vessels sprout from an existing vascular bed. This is a prerequisite for tumor growth beyond a certain size and for metastasis formation. Tumors produce a number of cytokines. The development of in vivo bio-assays for angiogenesis and in vitro analysis of endothelial cells permits characterization of these different cytokines concerning their role in angiogenesis. Some cytokines act mitogenically on endothelial cells, others have chemotactic activity or induce tube formation and some have multiple functions. A few factors are chemotactic for macrophages, infiltrating the tumor and secreting further angiogenic cytokines. Another important role in the process of angiogenesis is played by the extracellular matrix. Proteases secreted by all cell types involved (tumor cell, endothelial cell, macrophages) degrade the extracellular matrix, thereby releasing and activating angiogenic factors sequestered in the extracellular matrix. Thus tumor cells, macrophages and extracellular matrix release cytokines which together act on endothelial cells, resulting in the growth and infiltration of new blood vessels into the tumor.

Dexamethasone inhibits glioma-induced formation of capillary like structures in vitro and angiogenesis in vivo.

Dexamethasone is used frequently in brain tumor therapy of patients. In animal models it is known to inhibit the angiogenesis of solid tumors. We addressed the question, if this is also true in brain tumors. C6 malignant glioma and 9L gliosarcoma cells were implanted into rat-brains. Dexamethasone 3 mg/kg/d intraperitoneal increased the survival compared to saline treated controls. The tumors size and the vascular density were smaller in the dexamethasone groups in both models. In vitro dexamethasone inhibited the growth of the C6 cells but not of 9L cells. Thus the growth inhibition of brain tumors in vivo appeared to be mediated partly by direct growth inhibition of tumor cells in C6 cells but additionally by antiangiogenesis in both tumor models. Several in vitro models were used to address the mechanisms of antiangiogenesis. There was no effect of dexamethasone on the proliferation of central nervous endothelial cells and no effect on the formation of capillary like structures on matrigel. Dexamethasone inhibited, however, the formation of capillary like structures in a coculture model with glioma cells in vitro. Surprisingly, progesterone had the same effect in this model. The in vitro effect was mediated via glucocorticoid receptors since receptor antagonists could inhibit it. The primary target appeared to be the tumor cell because only this cell had the complete set of receptors. These data show, that antiangiogenic therapeutic effects are possible by influencing primarily the tumor cell. This way of targeting might be of value for future developments of new strategies.

C6 cells cross-resistant to cisplatin and radiation.

Malignant gliomas are relatively resistant to radiation and chemotherapy. To investigate whether cisplatin (cis-diamminedichloroplatinum(II), CDDP) causes resistance we pretreated C6 cells with 10(-6) M CDDP for 24 hours and then tested their sensitivity in a colorimetric assay. Pretreated cells developed resistance to CDDP (resistance factor 2.0) and radiation (survival after 9 Gy 60Co: 36.4% +/- 5.5 versus 28.6% +/- 5.2, p = 0.005). Glutathione levels of pretreated cells were higher (51.7 +/- 13.8 ng/mg protein) than in wt cells (40.4 +/- 13.2, p = 0.029). Addressing the mechanisms we established 4 wild type subclones with different CDDP sensitivities. However, cross-resistance to CDDP (survival: 60.7% +/- 3.5 versus 7.2% +/- 0.5, respectively p = < 0.001) and radiation (29.7% +/- 2.6 versus 12.9% +/- 0.8, p = < 0.001) could also be induced in a subclone showing involvement of mutation. These data suggest that CDDP can induce resistance mediated via induced mutation and increased GSH levels.

Cisplatin induces radioprotection in human T98G glioma cells.

Malignant gliomas are often treated with cisplatin (cis-diamminedichloroplatinum(II), CDDP) and radiation but results remain unsatisfactory. To investigate whether CDDP induces radioresistance in glioma, T98G human glioblastoma cells were pretreated 5 times with 10(-6)M CDDP for 24 hours and then the sensitivity of wild type (wt) and pretreated cells towards radiation (9Gy 60Co) and CDDP was tested in a colorimetric assay (MIT). The growth rates of wt and pretreated cells were 1.8 +/-0.2 and 3.1 +/- 0.2 respectively (p = 0.000155) 216 hours post radiation. Pretreated cells also developed resistance to CDDP (resistance factor 2.35). Glutathione (GSH) which potentially mediates resistance to both treatments was measured. Incubation for 6 hours with 10(-5) M CDDP increased GSH levels by a factor of 2.28 (p < 0.0001). However, neither basal nor increased levels differed between wt and pretreated cells. These data show that CDDP pretreatment can induce resistance against radiation and CDDP independently of GSH.