Activation of Endothelial Cells (EC) Leads to Release of Inhibitory Endothelial Microparticles (EMP) and Decreased Intracellular ADAMTS13 Activity.

BACKGROUND: Deficiency of the von Willebrand factor (vWF)-cleaving protease ADAMTS13 is known to be involved in the pathogenesis of idiopathic thrombotic thrombocytopenic purpura (TTP). A recent study has shown that in addition to liver, endothelial cells (EC) are a site of production of ADAMTS13 [D. Shang et al, Blood online 2006, prepub.]. Perturbation of EC has been implicated as the primary lesion in TTP and it has been demonstrated that markers of EC activation are increased in TTP. We previously documented elevated EMP positive for unusually large vWF (ULvWF) in TTP patients, and that EMP may inhibit ADAMTS13. This report confirms those findings by a more quantitative fluorogenic assay of ADAMTS13 activity, and extends them to the activities of whole endothelial cells (EC) and their interaction with EMP.

METHODS. Renal and brain microvascular EC (RMVEC, BMVEC) were cultured to confluency, then treated for 24 hr with TNF-α at 10 ng/mL to induce activation, or growth factor deprivation (GFD) to induce apoptosis. EMP were pelleted from the supernatants at 15,000×g, 30min. The parent cells were processed as described by Shang et al for assay of ADAMTS13 activity, which we assayed by the method of Kokame et al [Br J Haematol 129(1):93–100, 2005]. EMP were measured in flow cytometry by CD31+/CD42−, CD62E+ and CD62E+/vWF+. Concentrations are expressed in millions/mL. To measure effect of EMP on plasma ADAMTS13 activity, particle-free plasma (PFP) was incubated with 1×10³ to 1×10⁶ EMP/mL for 15min to 3hr, then centrifuged 1hr at 15,000×g to remove EMP, and activity remaining was assayed, as was the pelleted EMP.

RESULTS. Treatment with TNF-α caused a large increase in CD62E+ and CD62E+/vWF+ EMP above controls (3.6 ±1.3 vs. 0.33 ±0.12 for ctl; and 1.86 ±0.8 vs. 0.14 ±0.07 for ctl, respectively, p<0.01 for both). EMP from apoptotic MVEC (GFD-treated) exhibited a quite different profile, with CD31+ EMP accounting for 79% of all EMP (1.8 ±0.7 vs. 0.45 ±0.17 induced by TNF-α). Addition of EMP from TNF-α activated cells depressed ADAMTS13 activity in PFP by 35% ±14% in a time- and dose-dependent manner; maximal effect was at 1×10⁶ EMP/mL for 3hr. EMP from untreated cultures compared at equal concentration had weaker effect, depressing activity by 12% ±5%. EMP from apoptotic cells (GFD-treated) were also weaker than by TNF-α activation, causing 15% ±3% inhibition of activity in PFP. Next, we assayed ADAMTS13 activity of the parent MVEC, lysed according to method of Shang cited above. We found that TNF-α treated MVEC exhibited 65% ±20% less activity than quiescent MVEC, p<0.01.

CONCLUSION. These findings further indicate that upon activation, endothelial cells release and lose a significant fraction of ADAMTS13 activity. The EMP released are shown to be inhibitory to ADAMTS13. Hence, EMP may play a role in TTP by decreasing ADAMTS13 activity or availability.