NF-kappaB-independent induction of endothelial cell apoptosis by the vascular disrupting agent DMXAA.

BACKGROUND: DMXAA (5,6-dimethylxanthenone-4-acetic acid; AS1404), a vascular disrupting agent currently in clinical trials, induces tumour endothelial cell apoptosis in vivo in mice and in cancer patients. DMXAA activates NF-kappaB in many different cell types. In this study, whether DMXAA-induced endothelial cell apoptosis was NF-kappaB dependent was determined. MATERIALS AND METHODS: HUVEC endothelial and T24 endothelial-like cells were treated with DMXAA and apoptosis was measured by terminal deoxynucleotidyl transferase biotin-dUTP nick end labelling (TUNEL). NF-kappaB activation was measured by electrophoretic mobility shift assays (EMSA). T24 cells were transfected with IkappaBalphaM, a mutant form of the IkappaBalpha gene which cannot be phosphorylated and degraded, hence preventing NF-kappaB expression. RESULTS: No NF-kappaB up-regulation was detected in apoptotic HUVEC treated with DMXAA. The IkappaBalphaM-transfected T24 cells showed similar apoptotic responses to those of parental cells. CONCLUSION: The DMXAA-induced apoptosis is neither mediated by, nor inhibited by, the expression of the NF-kappaB pathway.

A comparison of the ability of DMXAA and xanthenone analogues to activate NF-kappaB in murine and human cell lines.

DMXAA (5,6-dimethylxanthenone-4-acetic acid), the most potent of a series of xanthenone (XAA) analogues developed in this laboratory, is currently undergoing combination clinical trials as an antivascular agent for cancer treatment. XAAs have a complex mode of action, and in vitro assays that are predictive of in vivo antitumor activity have been difficult to develop. In this study, we have utilized a series including XAA, DMXAA, and mono-substituted XAA derivatives to determine firstly whether in vitro NF-kappaB activation of mouse cell lines predicts for the in vivo antitumor potential of this class of agents, and secondly whether the relative activity of these analogues is similar in murine and human cell lines. Electromobility shift assays were used to measure NF-kappaB activation in murine HECPP endothelial and 70Z/3 pre-B cells, and in human HPLNEC.B3 endothelial and Raji B-lymphoma cells. A significant correlation was obtained between NF-kappaB activation in HECPP cells by a series of XAA analogues at 100 microg/ml (r = 0.78, p = 0.008) and at 300 microg/ml (r = 0.75, p = 0.01) and the amount of hemorrhagic necrosis induced in Colon 38 tumors. Different structure-activity relationships were observed in human and murine cell lines. 8-MeXAA, which was inactive in HECPP and 70Z/3 murine cell lines, showed similar NF-kappaB activation to DMXAA in human HPLNEC.B3 cells and Raji B-lymphoma cells. These results suggest that the receptor protein(s) in human cells that mediate the human response may have a lower stringency to that for murine cells. We also noted differences in the dose-response relationships for NF-kappaB activation between lymphoid and endothelial lines that were species independent. With increasing concentrations of DMXAA, NF-kappaB activation in both murine and human lymphoid lines showed a reproducible fluctuation, while in endothelial lines, the intensity of NF-kappaB activation was relatively constant above a threshold concentration. The results demonstrate interspecies differences in the NF-kappaB response to XAA analogues, and may also reflect the complex nature of NF-kappaB regulation.