Inhibition of thromboxane synthase activity improves glioblastoma response to alkylation chemotherapy

Abstract

Thromboxane synthase (TXSA), an enzyme of the arachidonic acid metabolism, is upregulated in human glial tumors and is involved in glioma progression. Here, we analyzed the in vitro and in vivo effects of pharmacological inhibition of TXSA activity on human glioblastoma cells. Furegrelate, a specific inhibitor of TXSA, significantly inhibited tumor growth in an orthotopic glioblastoma model by inducing proapoptotic, antiproliferative, and antiangiogenic effects. Inhibition of TXSA induced a proapoptotic disposition of glioma cells and increased the sensitivity to the chemotherapeutic agent 1,3-bis(2-chloroethyl)-1-nitrosourea, significantly prolonging the survival time of intracerebral glioma-bearing mice. Our data demonstrate that the targeted inhibition of TXSA activity improves the efficiency of conventional alkylation chemotherapy in vivo. Our study supports the role of TXSA activity for the progression of malignant glioma and the potential utility of its therapeutic modulation for glioma treatment.

Figure 1

In vitro effects of the specific TXSA inhibitor furegrelate on the human U87 glioblastoma cell line. (A) Increasing concentrations of furegrelate reduced the cell growth of U87. (B) Cotreatment of U87 with furegrelate and BCNU leads to a robust synergistic apoptotic effect in vitro indicating a sensitizing effect of furegrelate. Tumor cells were incubated with indicated concentrations and combinations of furegrelate and/or BCNU for 48 hours. DNA fragmentation was determined in cellular lysates (bars) and culture supernatant (o). No significant amounts of DNA fragments could be detected in the cell culture supernatant indicating that the effects were a consequence of apoptotic cell death and not of cellular necrosis. All values shown are mean ± SD. (C) Cotreatment with furegrelate and BCNU lead to a synergistic decrease in the surviving fraction of U87 tumor cells as determined by a colony formation assay. Values were calculated as mean ± SD of triplicate determinations.

Figure 2

The local intracerebral infusion (i.c.) of 2 mg/kg per day of furegrelate is more effective than 0.5 mg/kg per day on established human U87 glioblastoma xenografts in nude mice. (A) The local delivery of 2 mg/kg per day of furegrelate reduced the tumor growth of U87 human glioblastoma xenografts by 73.1% (*P < .05) as assessed by histologic volume measurement 21 days after treatment was initiated. Treatment with 2 mg/kg per day of furegrelate induced (B) a significantly increased apoptosis index (2.9-fold, *P < .005). Furthermore, treated tumors displayed (C) a significantly reduced intratumoral microvessel density (30.4%, *P < .05) and (D) a significant reduction of tumor cell proliferation (24.1%, *P < .01).

Figure 3

Histologic examination after 21 days of treatment with the local intracerebral microinfusion of 2 mg/kg per day of furegrelate. Representative examples of U87-derived tumors grown in the brain of animals that received PBS i.c. (A–D) or 2 mg/kg per day of furegrelate locally (A′–D′). (A, A′) Frozen sections stained with hematoxylin and eosin. (B, B′) Apoptotic tumor cells as assessed by immunoreactivity for cleaved caspase 3. (C, C′) Microvessel density as assessed by immunoreactivity for the endothelial cell marker CD31. (D, D′) Proliferating tumor cells as assessed by immunoreactivity for Ki67. Scale bar, 100 µm.

Figure 4

Kaplan-Meier survival curve of human U87 glioblastoma-bearing nude mice. The local intracerebral microinfusion (i.c.) of furegrelate at 2 mg/kg per day from day 8 to 36 in combination with a BCNU chemotherapy administered i.p. at 15 mg/kg per day from day 10 to 14 further increased the survival benefits of BCNU therapy (P < .0001).