Antiangiogenic Antithrombin

Confluent passage-2 HUVECs in Falcon T12.5 flasks grown in the presence or absence of VEGF-165 (10ng/mL) (R&D systems) were treated with human native, cleaved or latent antithrombins (20 5g/mL) for 24 hours in 3 mL M-199 with 2% heat-inactivated FBS and antibiotics. Following this treatment, total RNA was extracted from the cells by the Trizol method (Invitrogen). The total RNA was subjected to one cycle of linear RNA amplification using the MessageAmp aRNA kit according to the manufacturers instructions (Ambion, Austin, TX). The quality of the antisense RNA was verified on a denaturing agarose gel. Only samples showing a distribution of sizes from 250-5500 nt with a peak centered at 1000-1500 nt were used to generate the test cDNA samples for the subsequent array experiments. DNA labeling and hybridizations were performed essentially as described (Pollack et al., 1999, Nat. Genet.), with slight modifications. Briefly, the test endothelial cell antisense RNA samples together with Universal Human Reference RNA (Stratagene, Cedar Creek, TX) were used to generate Cy3/Cy5 (Amersham Biosciences) labeled cDNA for array hybridization on the Stanford 43K human cDNA microarray (www.microarray.org/sfgf). The Stanford microarray used in this study consists of 18,416 named genes with UniGene symbol, 4,145 ESTs with known function, 19,365 ESTs with unknown function and ~1,000 repeated spots as internal controls. Hybridized arrays were scanned on a GenePix 4000B scanner (Axon Instruments), and fluorescence ratios (test/reference) calculated using the Stanford Microarray Database software (available at