Ability of Polyphosphate and Nucleic Acids to Trigger Blood Clotting: Some Observations and Caveats

Stephanie A. Smith, Joshua M. Gajsiewicz, James H. Morrissey

Polyphosphate plays several roles in coagulation and inflammation, while extracellular DNA and RNA are implicated in thrombosis and as disease biomarkers. We sought to compare the procoagulant activities of polyphosphate versus DNA or RNA isolated from mammalian cells. In a recent study, we found that much of the procoagulant activity of DNA isolated from mammalian cells using Qiagen kits resisted digestion with nuclease or polyphosphatase, and even resisted boiling in acid. These kits employ spin columns packed with silica, which is highly procoagulant. Indeed, much of the apparent procoagulant activity of cellular DNA isolated with such kits was attributable to silica particles shed by the spin columns. Therefore, silica-based methods for isolating nucleic acids or polyphosphate from mammalian cells are not suitable for studying their procoagulant activities. We now report that polyphosphate readily co-purified with DNA and RNA using several popular isolation methods, including phenol/chloroform extraction. Thus, cell-derived nucleic acids are also subject to contamination with traces of cellular polyphosphate, which can be eliminated by alkaline phosphatase digestion. We further report that long-chain polyphosphate was orders of magnitude more potent than cell-derived DNA (purified via phenol/chloroform extraction) or RNA at triggering clotting. Additional experiments using RNA homopolymers found that polyG and polyI have procoagulant activity similar to polyphosphate, while polyA and polyC are not procoagulant. Thus, the procoagulant activity of RNA is rather highly dependent on base composition.