Search

Heparin (H), heparan sulfate (HS), and related glycosaminoglycans can inhibit cancer cell invasion, possibly due to their ability to interact with vascular growth factors, adhesion molecules, endoglycosidases, and signaling proteins, in addition to the well-known effects on the clotting system. We evaluated the antitumor activity of a series of semisynthetic sulfaminoheparosan sulfates (SAHSs) with different degree and distribution of sulfates, obtained by chemical modifications of the E. coli K5 polysaccharide, namely type A, B, and C compounds. B16-BL6 melanoma cells (10 5 cells/mouse) were injected intravenously (i.v.) in a lateral tail vein of C57BL6 mice at a dose of 0.5 mg/ mouse together with test compounds. Tumor lung nodules were significantly reduced as compared with controls only by H (95.5 +/- 1.0% inhibition), SAHS-2 (84.2 +/- 5.0% inhibition), and SAHS-4 (91.1 +/- 4.2% inhibition), among compounds tested. SAHS-2 and SAHS-4 are type B compounds, with a sulfate/carboxylate ratio similar to that of H. A typical mammalian HS showed only 54.8% inhibition. Supersulfated low-molecular-weight heparin and heparan sulfate (ssLMWH and ssLMWHS) showed an activity similar to that of unfractionated compounds. H and SAHS-4 inhibited dose dependently B16-BL6 lung colonies, with IC-50 values of 0.05 and 0.1 mg/mouse, respectively. The relationship with ex vivo anticoagulant potency was evaluated by activated partial thromboplastin time (aPTT) on mouse plasma at different time intervals after i.v. injection (0.1 to 0.5 mg/mouse) of the compound. H showed a dose-dependent anticoagulant activity lasting up to 2 hours, whereas SAHS-4 showed a potent anticoagulant effect only at a dose of 0.5 mg/mouse. Accordingly, H but not SAHS-4 consistently inhibited B16-BL6 lung colonies when given 1 hour before tumor cells. SAHS-4 derivatives, with different size and/or affinity depleted of AT binding sites, showed an inhibitory effect on B16-BL6 melanoma similar to that of SAHS-4, suggesting that the greater antitumor effect of H was not due to AT-mediated inhibition of blood clotting. Interactions with other blood inhibitors, such as heparin cofactor II or tissue factor pathway inhibitory protein cannot be ruled out. The better effect of H may be due to persistence in the circulation and/or ability to inhibit tumor neoangiogenesis.

B16 melanoma cells selected in mice for liver-specific metastasis (B16-LS9) overexpress a constitutively active form of the hepatocyte growth factor/scatter factor receptor (HGF/SFr), the product of the c-met proto-oncogene. HGF/SF can affect both invasion and growth of receptive cells. In fact, we show that overexpression of c-met in B16-LS9 cells results in a higher inducibility of two different proteolytic activities (uPA and gelatinase), in correlation with a stronger invasive and motility response to HGF/SF treatment. However, HGF/SF treatment inhibits growth of B16 cells, which might appear in contradiction with the observation that c-met overexpression and constitutive activation seems to be required for efficient liver colonization. However, this apparent discrepancy is resolved by the finding that liver-derived, but not lung-derived factor(s), can efficiently rescue B16-LS9 cells from the growth inhibitory effects of HGF/SF, while not changing their motility response. Therefore, overexpression of c-met in B16-LS9 cells might give a specific advantage in liver colonization, because specifically at this site B16-LS9 cells can take full advantage of the positive effects exerted by HGF/SF stimulation on motility and invasion, while the negative effects on growth are counteracted by other paracrine factor(s).