Induction of thrombospondin-1 partially mediates the anti-angiogenic activity of dexrazoxane.

BACKGROUND: Considerable interest lies in the identification of novel anti-angiogenic compounds for cancer therapy. We have investigated whether dexrazoxane has anti-angiogenic properties and if so, the mechanism of the inhibition. METHODS: The phenotypic effects of dexrazoxane on endothelial cell behaviour was investigated both in vitro using human umbilical vein endothelial cells (HUVECs) in cell proliferation, migration, cell cycle and aortic ring assays; and in vivo using the mouse angiogenesis subcutaneous sponge assay. Custom angiogenesis pathway microarrays were used to identify differentially expressed genes in endothelial cells after treatment with dexrazoxane vs a control. The differentially expressed genes were validated using real-time RT-PCR and western blotting; and the functional effect of one induced gene was confirmed using siRNA technology. RESULTS: Treatment of endothelial cells with dexrazoxane resulted in a dose-response inhibition of cell growth lasting for up to 5 days after a single dose of the drug. Dexrazoxane was inhibitory in the aortic ring tube forming assay and strongly anti-angiogenic in vivo in the rodent subcutaneous sponge model. The anti-angiogenic effect in the sponge was seen after systemic injection into the tail vein as well as after direct injection of dexrazoxane into the sponge. Treatment of microvascular endothelial cells in vitro with subtoxic doses of dexrazoxane stimulated thrombospondin-1 (THBS-1) secretion. Knockdown of THBS-1 with siRNA removed the angiogenesis inhibition effect of dexrazoxane, which is consistent with the anti-angiogenic and vascular normalising properties of the drug being principally mediated by THBS-1. CONCLUSION: We show that dexrazoxane administered in small repeated doses is strongly anti-angiogenic and that this activity is mediated by induction of the anti-angiogenic THBS-1 in endothelial cells.

Effects of rosiglitazone and linoleic acid on human preadipocyte differentiation.

BACKGROUND: Peroxisome proliferator activated receptor gamma (PPARgamma) is a ligand-activated transcription factor known to be central to both adipose tissue development and insulin action. Growth of adipose tissue requires differentiation of preadipocytes with acquisition of specific cellular functions including insulin sensitivity, leptin secretion and the capacity to store triglyceride. Dietary fatty acids and members of the thiazolidinedione class of compounds have been reported to influence adipogenesis at the transcriptional level. Here, we compare the effects of a dietary fatty acid, linoleic acid, and a thiazolidinedione, rosiglitazone, on biochemical and functional aspects of human preadipocyte differentiation in vitro. MATERIALS AND METHODS: Human omental and subcutaneous preadipocytes were subcultured 2-3 times and subsequently differentiated for 21 days in the presence of either linoleic acid or rosiglitazone. Differentiation was assessed using a number of biochemical and functional parameters. RESULTS: Omental and subcutaneous preadipocytes differentiated in the presence of linoleic acid showed marked cytoplasmic triacylglycerol accumulation however, no biochemical markers of differentiation (LPL expression, G3PDH gene expression and enzyme activity and leptin expression or secretion) were detected. In contrast, treatment of these cells with rosiglitazone induced full biochemical differentiation as judged by all markers assessed, despite comparatively little lipid accumulation. The rosiglitazone effects were subcutaneous depot-specific. Cells treated with linoleic acid showed decreased glucose uptake cf rosiglitazone-treated cells. A luciferase reporter assay demonstrated that rosiglitazone potently activates h-peroxisome proliferator activated receptor gamma while linoleic acid had no effect. CONCLUSIONS: These studies demonstrate that (a) human preadipocytes have the potential to accumulate triacylglycerol irrespective of their stage of biochemical differentiation; (b) while omental preadipocytes are refractory to biochemical differentiation in vitro, they are able to accumulate triacylglycerol; and (c) rosiglitazone and linoleic acid may exert their effects via different biochemical pathways.

Genetic and physical interactions between Microphthalmia transcription factor and PU.1 are necessary for osteoclast gene expression and differentiation.

The microphthalmia transcription factor (MITF), a basic-helix-loop-helix zipper factor, regulates distinct target genes in several cell types. We hypothesized that interaction with the Ets family factor PU.1, whose expression is limited to hematopoietic cells, might be necessary for activation of target genes like tartrate-resistant acid phosphatase (TRAP) in osteoclasts. Several lines of evidence were consistent with this model. The combination of MITF and PU.1 synergistically activated the TRAP promoter in transient assays. This activation was dependent on intact binding sites for both factors in the TRAP promoter. MITF and PU.1 physically interacted when coexpressed in COS cells or in vitro when purified recombinant proteins were studied. The minimal regions of MITF and PU.1 required for the interaction were the basic-helix-loop-helix zipper domain and the Ets DNA binding domain, respectively. Significantly, mice heterozygous for both the mutant mi allele and a PU.1 null allele developed osteopetrosis early in life which resolved with age. The size and number of osteoclasts were not altered in the double heterozygous mutant mice, indicating that the defect lies in mature osteoclast function. Taken in total, the results afford an example of how lineage-specific gene regulation can be achieved by the combinatorial action of two broadly expressed transcription factors.