Development of oligosaccharides as anti-angiogenic agents

Proc Amer Assoc Cancer Res, Volume 46, 2005 3035 Several angiogenic cytokines are dependent on heparan sulfate (HS) for their biological activity. The prototypic HS-dependent growth factor is FGF2 and here we explore the relevance of FGF2 and HS proteoglycans (HSPG) to human ovarian cancer vasculature and perform the first detailed in vivo analysis of the anti-angiogenic potential of size defined oligosaccharides: We show that the HSPG, syndecan 3, is aberrantly expressed by the tumor vasculature. Although anti-HS antibodies detect HS that has the ability to bind FGF2 throughout ovarian cancer tissue, a novel FGFR1-IIIc-alkaline phosphatase probe, which detects heparan sulfate that has the capacity to form a trimeric complex with HS and FGF2, principally binds to the tumor endothelium implying that this tissue is the predominant source of biologically active (FGF2-activating) HS. Further studies with RT-PCR and RNA in situ hybridization confirm the presence of FGFR1-IIIc in the ovarian cancer tissue that is, in particular, expressed by the endothelium. Taken in conjunction with other data confirming that FGF2 is present in ovarian cancer tissue, we show for the first time that the entire extra-cellular signaling apparatus for FGF2 is present in ovarian cancer endothelium, highlighting the potential importance of this cytokine as a target for anti-angiogenic agents. We wanted to investigate the potential of oligosaccharides to inhibit angiogenesis caused by FGF2 and therefore we now report the first detailed in vivo studies of the inhibition of angiogenesis by size fractionated heparin oligosaccharides. The latter were prepared by size exclusion gel chromatography of commercially available low molecular weight heparin and species containing 4, 6, 8, 10, 12 or 14 saccharide residues were administered subcutaneously at 20mg/kg/day for up to 3 weeks. Four models that were progressively less dependent on FGF2 were investigated: 1) Using a sponge model of angiogenesis, where a subcutaneous sponge is injected with FGF2 100ng/day, we showed that heparin octa- and deca-saccharides completely inhibited FGF2 induced angiogenesis, estimated through microvascular density. 2) A similar effect was seen in endometrial carcinoma cells, transduced to over-express FGF2, that were contained in a hollow fibers placed subcutaneously. 3) When the latter experiment was repeated with H460 lung carcinoma cells in hollow fibers in vivo, angiogenesis was promoted by a mixture of angiogenic cytokines. In this case octasaccharides reduced the microvessel density by 70%. 4) Lastly when H460 lung carcinoma cells were grown as xenograft tumors, the doubling time increased from 13 days in control animals to 17 days in octasaccharide treated animals. These are the first detailed studies of size defined oligosaccharides in models of angiogenesis and show that heparin octasaccharides have anti-angiogenic and anti-tumor effects in vivo.