Background/Purpose:  Antiphospholipid antibodies, especially those targeting beta-2-glycoprotein I (β₂GPI), have been shown to amplify thrombosis in mice.  However, most published models have relied on endothelial injury, induced by either laser or pinch injury.  Here, we have characterized a model of deep vein thrombosis (DVT) in which flow through the inferior vena cava (IVC) is restricted without causing complete occlusion, and in which the endothelium is not specifically damaged.  This model results in a thrombus that recapitulates the histologic characteristics of human DVT.

Methods:   Total IgG fractions were prepared from either healthy volunteers or two different primary antiphospholipid syndrome (APS) patients, both with high-titer anti-β₂GPI IgG.  C57BL/6 wild-type mice were treated with two doses (500 µg each) of either control or APS IgG by intraperitoneal injection, 48 hours apart.  Around the time of the second treatment, a laparotomy was performed, and a ligature was fastened around the IVC over a blunted 30-gauge needle (which served as a spacer).  After removal of the spacer, the abdomen was closed, and the mouse was allowed to recover.  The resulting stenosis causes an 80-90% reduction in IVC blood flow.  Some mice were additionally treated with infusion of DNase (Pulmozyme/dornase alfa) immediately after surgery.  DVT formation was assessed at 6 and 48 hours after surgery.

Results:   APS IgG injection resulted in circulating anti-β₂GPI IgG levels of 10-20 SGU (standard IgG units), in contrast to control IgG, which gave undetectable levels.  Importantly, mice treated with APS IgG were significantly more likely to form IVC thrombi than mice treated with control IgG.  At 6 hours, 40% of control mice formed thrombi, as compared to >90% of APS mice.  At 48 hours, these numbers were 50-60% for control mice, and >90% for APS mice.  By both histology and western blotting, the thrombi were rich in citrullinated histone H3 (cit-H3; a marker of neutrophil extracellular traps/NETs),  There was also a significant increase in circulating cell-free DNA in the APS mice, as compared to controls, at the 6-hour time point.  Treatment with DNase was highly effective in preventing thrombus formation at 6 hours, reducing the percentage of clots in APS mice from >90% to ~33%.  Further supporting a mechanistic role for NETs, anti-β₂GPI IgG promoted NET release from C57BL/6 neutrophils in vitro.  Of note, a spontaneous model of APS (NZW x BXSB F₁) was also associated with >90% DVT formation following IVC stenosis; neutrophil characterization and treatment experiments are underway in these mice.

Conclusion: IgG isolated from APS patients increase the frequency of thrombus formation in a model of DVT that recapitulates the histology of human thrombi.  This increase can be abrogated by treatment with DNase, suggesting a role for extracellular DNA in thrombus formation.  Further, cit-H3 was abundant in non-nuclease-treated thrombi.  Overall, these data point to a role for NETs as perpetuators, and possibly initiators, of the prothrombotic phenotype in APS.  This line of investigation has the potential to suggest new, non-anticoagulant approaches for the treatment of APS.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/dnase-treatment-protects-against-dvt-formation-in-a-mouse-model-of-antiphospholipid-syndrome/