Synthesis and biological evaluation of 1-(benzenesulfonamido)-2-[5-(N-hydroxypyridin-2(1H)-one)]acetylene regioisomers: a novel class of 5-lipoxygenase inhibitors.

A hitherto unknown class of linear acetylene regioisomers were designed such that a SO(2)NH(2) group was located at the ortho-, meta-, or para-position of the acetylene C-1 phenyl ring, and a N-hydroxypyridin-2(1H)-one moiety was attached via its C-5 position to the C-2 position on an acetylene template (scaffold). All three regioisomers inhibited 5-lipoxygenase (5-LOX), where the relative potency order was 2-SO(2)NH(2) (IC(50)=10 microM) >3-SO(2)NH(2) (IC(50)=15 microM) >4-SO(2)NH(2) (IC(50)=68 microM) relative to the reference drug nordihydroguaiaretic acid (NDGA; IC(50)=35 microM). The 2-SO(2)NH(2) regioisomer (ED(50)=86.0mg/kg po) exhibited excellent oral anti-inflammatory (AI) activity that was more potent than aspirin (ED(50)=128.9 mg/kg) and marginally less potent than ibuprofen (ED(50)=67.4 mg/kg). The N-hydroxypyridin-2(1H)one moiety provides a novel pharmacophore for the design of cyclic hydroxamic mimetics capable of chelating 5-LOX iron for exploitation in the design of 5-LOX inhibitory AI drugs.

Technical hurdles in a pilot clinical trial of combined B7-2 and GM-CSF immunogene therapy for glioblastomas and melanomas.

OBJECTIVE: Malignant glioblastomas and melanomas continue to have a dismal prognosis despite advances in conventional therapy. This has led to investigations of novel treatment strategies including immunogene therapy. We report a pilot clinical trial of combined B7-2 and GM-CSF immunogene therapy for gliomas and melanomas and discuss technical hurdles encountered. METHODS: Patients with recurrent malignant gliomas or medically refractory melanomas were vaccinated with irradiated autologous tumor cells transduced with B7-2 and GM-CSF genes using a retroviral vector. Patients were monitored for toxicity, inflammatory/immune reactions, and clinical status. RESULTS: Vaccine preparation was attempted from 116 malignant glioma and 32 melanoma specimens. Adequate vaccines could only be prepared for five glioblastoma and three melanoma patients. Six patients (three recurrent glioblastomas and three melanomas) were actually vaccinated. Minor toxicities included flu-like symptoms (3/6), injection site erythema (4/6), and asymptomatic elevations in liver enzymes (3/6). Most patients showed evidence of an inflammatory response but specific anti-tumor immunity was not demonstrated. All six patients have died, although three patients with minimal residual disease at treatment had prolonged recurrence-free intervals after vaccination. CONCLUSIONS: Combined B7-2 and GM-CSF immunogene therapy for glioblastomas and melanomas using autologous tumor cells has many technical pitfalls hindering large scale application and evaluation. As a result, this pilot study was too limited to draw meaningful conclusions regarding safety or anti-tumor immunity. While immunotherapy has been promising in pre-clinical studies, alternate strategies will be required to bring these benefits to patients.

TRAIL inhibits tumor growth but is nontoxic to human hepatocytes in chimeric mice.

Tumor necrosis factor (TNF) family ligand TNF-alpha and Fas ligand (FasL) can trigger apoptosis in solid tumors, but their clinical usage has been limited by hepatotoxicity. TNF-related apoptosis-inducing ligand (TRAIL) is a newly identified member of the TNF family, and its clinical application currently is under a similar debate. Here, we report a recombinant soluble form of human TRAIL (114 to 281 amino acids) that induces apoptosis in tumor cells but not human hepatocytes. We first isolated human hepatocytes from patients and showed that the human hepatocytes expressed Fas but no TRAIL death receptor DR4 and little DR5 on the cell surface. Antibody cross-linked FasL, but not TRAIL, triggered apoptosis of the human hepatocytes through cleavage of caspases. We then examined TRAIL hepatotoxicity in severe combined immunodeficient/Alb-uPA chimeric mice harboring human hepatocytes. Intravenous injection of FasL, but not TRAIL, caused apoptotic death of human hepatocytes within the chimeric liver, thus killing the mice. Finally, we showed that repeated intraperitoneal injections of TRAIL inhibited intraperitoneal and subcutaneous tumor growth without inducing apoptosis in human hepatocytes in these chimeric mice. The results indicate that the recombinant soluble human TRAIL has a profound apoptotic effect on tumor cells but is nontoxic to human hepatocytes in vitro and in vivo.

TRAIL triggers apoptosis in human malignant glioma cells through extrinsic and intrinsic pathways.

Many malignant glioma cells express death receptors for tumor necrosis factor-related apoptosis-inducing ligand (TRAIL), yet some of these cells are resistant to TRAIL. Here, we examined signaling events in TRAIL-induced apoptosis and searched for therapeutic agents that could overcome TRAIL resistance in glioma cells. TRAIL induced apoptosis through death receptor 5 (DR5) and was mediated by caspase-8-initiated extrinsic and intrinsic mitochondrial pathways in sensitive glioma cell lines. TRAIL also triggered apoptosis in resistant glioma cell lines through the same pathways, but only if the cells were pretreated with chemotherapeutic agents, cisplatin, camptothecin and etoposide. Previous studies suggested that this was due to an increase in DR5 expression in wild-type TP53 cells, but this mechanism did not account for cells with mutant TP53. Here, we show that a more general effect of these agents is to downregulate caspase-8 inhibitor c-FLIP(S) (the short form of cellular Fas-associated death domain-fike interleukin-1-converting enzyme-inhibitory protein) and up-regulate Bak, a pro-apoptotic Bcl-2 family member, independently of cell's TP53 status. Furthermore, we showed that TRAIL alone or in combination with chemotherapeutic agents, induced apoptosis in primary tumor cultures from patients with malignant gliomas, reinforcing the potential of TRAIL as an effective therapeutic agent for malignant gliomas.

Cisplatin down-regulation of cellular Fas-associated death domain-like interleukin-1beta-converting enzyme-like inhibitory proteins to restore tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis in human melanoma cells.

PURPOSE: Many melanoma cell lines and primary cultures are resistant to tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-induced apoptosis. In this study, we investigated the molecular mechanisms that control melanoma cell resistance and searched for chemotherapeutic drugs that could overcome the TRAIL resistance in melanoma cells. EXPERIMENTAL DESIGN: We examined 21 melanoma cell lines and 3 primary melanoma cultures for their sensitivity to TRAIL-induced apoptosis, and then tested cisplatin, chemptothecin, and etoposide for their synergistic effects on TRAIL sensitivity in resistant melanoma cells. RESULTS: Of 21 melanoma cell lines, 11 showed various degrees of sensitivity to TRAIL-induced apoptosis through caspase-8-initiated cleavage of caspase-3 and DNA fragmentation factor 45. The remaining cell lines and primary cultures were resistant to TRAIL, but cisplatin, chemptothecin, and etoposide sensitized the resistant cell lines and primary cultures to TRAIL-induced apoptosis, which also occurred through the caspase-8-initiated caspase cascade. Of the two TRAIL death receptors (DR4 and DR5), melanoma cells primarily expressed DR5 on cell surface. Cisplatin treatment had no effects on cell surface DR5 expression or intracellular expression of Fas-associated death domain and caspase-8. Instead, cisplatin treatment down-regulated intracellular expression of the short form of cellular Fas-associated death domain-like interleukin-1beta-converting enzyme-like inhibitory protein (c-FLIP) and inhibited phosphorylation of the long form of c-FLIP. CONCLUSIONS: The results presented here indicate that cisplatin inhibits c-FLIP protein expression and phosphorylation to restore TRAIL-induced caspase-8-initiated apoptosis in melanoma cells, thus providing a new combined therapeutic strategy for melanomas.

Human autologous in vitro models of glioma immunogene therapy using B7-2, GM-CSF, and IL12.

BACKGROUND: Cancer immunogene therapy is based on vaccination with radiated, autologous tumor cells transduced with immunostimulatory genes. To help determine an optimal glioma immunogene therapy strategy, we stimulated lymphocytes with autologous human glioma cells transduced with B7-2 (CD86), granulocyte-macrophage colony-stimulating factor (GM-CSF), and/or interleukin-12 (IL12). METHODS: A human glioma-derived cell culture (Ed147.BT) was transduced with B7-2, GM-CSF, and/or IL12 using retroviral vectors. Autologous peripheral blood mononuclear cells (PBMC) were co-cultured with irradiated gene-transduced tumor alone or a combination of radiated wild type and gene-transduced cells. Peripheral blood mononuclear cells proliferation was determined by serial cell counts. Peripheral blood mononuclear cells phenotype was assessed by flow cytometry for CD4, CD8, and CD16. Anti-tumor cytotoxicity was determined by chromium-51 (51Cr) release assay. RESULTS: Peripheral blood mononuclear cells cell numbers all decreased during primary stimulation but tumor cells expressing B7-2 or GM-CSF consistently caused secondary proliferation. Tumors expressing B7-2 and GM-CSF or B7-2, GM-CSF, and IL12 consistently increased PBMC CD8+ (cytotoxic T) and CD16+ (natural killer) percentages. Interestingly, anti-tumor cytotoxicity only exceeded that of PBMC stimulated with wild type tumor alone when peripheral blood mononuclear cells were stimulated with both wild type tumor and B7-2/GM-CSF- (but not IL12) transduced cells. CONCLUSIONS: PBMC proliferation and phenotype is altered as expected by exposure to immunostimulatory gene-transduced tumor. However, transduced tumor cells alone do not stimulate greater anti-tumor cytotoxicity than wild type tumor. Only B7-2/GM-CSF-transduced cells combined with wild type produced increased cytotoxicity. This may reflect selection of tumor subclones with limited antigenic spectra during retrovirus-mediated gene transfer.

Human brain tumor cell culture characterization after immunostimulatory gene transfer.

OBJECTIVE: Immunogene therapy is a novel cancer treatment strategy based on vaccination with irradiated autologous tumor cells transduced with immunostimulatory genes. To characterize such cells before clinical applications, we studied a human glioma cell line (D54 MG) and early passage human glioma (Ed147.BT, Ed149.BT) and melanoma (Ed141.MEL) cultures after immunostimulatory gene transfer. METHODS: Granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-12 (IL-12), and B7-2 genes were retrovirally transferred to tumor cells. Gene expression before and after irradiation (200 Gy) was assessed by enzyme-linked immunosorbent assay (GM-CSF, IL-12) and flow cytometry (B7-2). Viability and clonogenicity were determined via trypan blue staining before and after irradiation. Growth rates were determined by serial cell counts. RESULTS: GM-CSF expression was high in GM-CSF-transduced (10.36-162.10 ng/10(6) cells/d preirradiation and 10.22-122.02 ng/10(6) cells/d postirradiation) but lower in B7-2/GM-CSF-transduced cultures (1.41-2.90 ng/10(6) cells/d preirradiation, 1.96-5.02 ng/10(6) cells/d postirradiation). IL-12 expression also was lower (1.30-2.10 ng/10(6) cells/d preirradiation, 0.47-1.70 ng/10(6) cells/d postirradiation). B7-2 expression was high (one- to two-logarithm increase in fluorescence) and unaffected by radiation. Postirradiation viability was initially high (94.20 +/- 8.46%, Day 1) but decreased rapidly (28.13 +/- 4.64%, Day 10). No cultures demonstrated evidence of clonogenicity (i.e., cell division) after 200-Gy irradiation. Growth rates were similar in wild-type and gene-transduced Ed141.MEL, Ed147.BT, and Ed149.BT. However, D54MG-IL-12 growth was slower than that of wild-type D54MG. CONCLUSION: GM-CSF, IL-12, and B7-2 genes can be transferred to human glioma and melanoma cell cultures efficiently by use of our retroviral vectors. Irradiation (200 Gy) does not significantly alter therapeutic gene expression. Irradiated cells remain viable for several days but cannot undergo further cell division. Early passage culture growth rates are not altered by therapeutic gene expression but are decreased by IL-12 in an immortalized cell line (D54MG). These results suggest that it is feasible to create vaccines with irradiated, autologous, genetically modified brain tumor cells.

Cytokine and cytokine receptor mRNA expression in human glioblastomas: evidence of Th1, Th2 and Th3 cytokine dysregulation.

Immunotherapies, although promising in preclinical studies, have not yet enhanced the survival of patients with glioblastomas. To further understand the immunobiology of glioblastomas in clinical settings, we examined 53 cytokine or cytokine receptor transcripts in 12 human glioblastomas and 6 human glioblastoma cell lines and correlated the findings with the degree of inflammation. Multi-probe RNase protection assays were used to examine Th1, Th2, and Th3 cytokine and cytokine receptor expression. Th2 [interleukin (IL)-6, leukemia inhibitory factor and oncostatin M] and Th3 (transforming growth factor-beta1, 2, 3) cytokine and their receptor transcripts were strongly expressed in almost all glioblastomas and glioma cell lines. Two other Th2 cytokine receptor subunit transcripts (IL-4Ralpha and IL-13Ralpha) were also commonly detected. In contrast, although Th1 cytokine receptors tumor necrosis factor (TNF) RI, interferon (IFN)-gammaRalpha, IFN-gammaRbeta, were detected, their cytokines (IFN-gamma, TNF-alpha, lymphotoxin-alpha) were not. Transcripts for IL-2 family cytokine (IL-2, IL-7, IL-9, IL-15) and receptors (IL-2Ralpha, IL-2Rbeta, gammac, IL-7Ralpha, IL-9Ralpha, IL15Ralpha) and IL-12 family cytokine (IL-12p40) and receptors (IL-12Rbeta1 and IL-12beta2) were essentially absent in both tumors and cell lines. Immunohistochemical methods showed sparse T lymphocyte infiltrates and numerous microglia in the glioblastomas. This pattern indicates an 'immunosuppressive status' in glioblastomas and could account for the failure of immunotherapy in such tumors.