Loss of Rnase1 Produces Coagulation Abnormalities in Mice

Ribonuclease 1 (RNase 1) is a secreted enzyme with robust and nonspecific endoribonucleolytic activity. It is present in all vertebrate species and is expressed ubiquitously in human tissue, but its biological function is not well understood. In recent years, however, clues to its activity have emerged. RNase 1 is most highly produced in vascular endothelial cells and has a pH optimum of 7.2, which suggests that its main activity is in the blood. Additionally, multiple publications have highlighted the ability of a homolog of RNase 1 (the more famous Ribonuclease A) to prevent or reduce RNA‐mediated thrombosis when administered to mice. These findings suggest a role for endogenous RNase 1 in modulation of the coagulation cascade, and we are evaluating this hypothesis in an Rnase1 knockout mouse system. Rnase1 ‐null mice are viable and fertile, exhibit similar lifespan to wild‐type mice, and exhibit no gross anatomical, behavioral, or metabolic abnormalities. In vitro coagulation assays do reveal shortened clotting times for Rnase1 ‐null plasma relative to wild‐type, with significantly shorter times for unstimulated plasma and using a modified prothrombin time assay. Yet, Rnase1 ‐null mice did not exhibit increased thrombin‐antithrombin complex formation in response to lipopolysaccharide challenge, did not bleed less than wild‐type mice in a tail‐vein bleeding test, and did not form thrombi more quickly than did wild‐type mice in a ferric‐chloride induced arterial thrombosis model. Mice that lack expression of intrinsic pathway coagulation factors, such as factor XI (fXI) and factor XII (fXII), also do not exhibit perturbed in vivo coagulation behavior despite prolongation of coagulation in vitro . Additionally, these factors are activated in vitro by RNA. Accordingly, and because Rnase1 ‐null plasma contains significantly more RNA than does wild‐type plasma, we are evaluating whether the loss of RNase 1 permits increased activation of the intrinsic coagulation pathway. Indeed, preliminary factor activity assays indicate that fXII is strongly activated in Rnase1 ‐null plasma. Immunoblotting data suggest the mechanism of this activation is autoactivation by conformational change of fXII, as can occur with RNA binding. Ongoing experiments are being carried out to confirm this mechanism in mice. The finding that RNase 1 affects coagulation in mice represents a new understanding of the biological function of this enzyme, and underscores our growing understanding of the importance of RNA in myriad biological functions. Support or Funding Information The authors are grateful to John Sheehan and his laboratory for technical and intellectual assistance with coagulation studies. Factor‐deficient plasma was a generous gift from Dave Gailani's laboratory (Vanderbilt University). Tim Hacker and the UW Cardiovascular Physiology Core facility provided technical assistance with in vivo coagulation assays, and we are grateful to Conner Feldman and Katharine Tippins for technical assistance with our mouse colony. This work was supported by the National Institutes of Health under award numbers T32GM008688 and R01CA073808.