Fig 2: Monoclonal antibody concentration–response curves in vitro. The concentration-dependent ADAMTS13 inhibitory activity of human (A) and mouse (B) antibodies was determined by incubating normal pooled plasma with 2-fold serial dilutions of the antibodies ranging from 25 µg/mL to 0.20 µg/mL in triplicate. The values of dose 0 control were also measured to calculate the residual ADAMTS13 activity of each antibody. The horizontal axis is the concentration of antibodies used in these tests. The vertical axis is the residual plasma ADAMTS13 activity relative to the values of dose 0 control.

Fig 3: Cleavage of von Willebrand factor (VWF) by matrix metalloproteinase‐13 (MMP‐13). A, SDS‐PAGE of cleaved VWF samples. MMP‐13 but not ADAMTS13 at a concentration of 1.5 µmol/L was able to cleave purified human VWF (0.2 mg/mL) after 2 hours at 37°C. Degradation products were analyzed by (i) 12% reducing SDS‐PAGE and (ii) overnight separation of high molecular weight multimers on 15% acrylamide gels at 4°C. B, Schematic representation of MMP‐13 cleavage sites on VWF. Sequence analysis of MMP‐13 cleavage sites revealed two N‐terminal to the A1 domain and one within the C8‐4 domain. The ADAMTS13 (a disintegrin and metalloprotease with thrombospondin motif member 13) cleavage site within the A2 domain is also marked for reference

Fig 4: WPB size modulates the recruitment of plasma vWF to the endothelial surface.(a) HUVECs were subjected to flow assays as indicated. When WPB exocytosis was stimulated, histamine (100 μmol/L) was supplemented to buffer and human normal pooled plasma. (b) At the end of the assays, cells were fixed and the extracellular vWF immuno-stained and imaged. Scale bar: 25 μm. (c) Diagram of the procedure in panels (a) and (b) illustrates the assay and what it measures. (d) vWF associated with the endothelial surface following the assays described in panel (a). The area of the vWF signal on the endothelial surface was measured from 36 fields of view (the data points displayed) per condition. Bars: medians and interquartile ranges; histamine stimulation in buffer and histamine stimulation coupled to plasma were compared to the unstimulated plus plasma condition; ****P < 0.0001 (Mann-Whitney test). (e) Luciferase- or vWF-siRNA treated HUVECs were subjected to the indicated flow assay (left) and vWF deposition on endothelial cells measured (right); ****P < 0.0001 (Mann-Whitney test). (f) vWF plasma recruitment after a brief treatment with recombinant ADAMTS13 (rADAMTS13) at the indicated concentrations; *P < 0.05; ***P < 0.001 (Mann-Whitney test). (g) HUVECs treated for 24 h with vehicle (DMSO, control), medium at pH 6.4 and supplemented with acetate or nocodazole (1 μg/mL) were processed for immunofluorescence. Scale bar: 5 μm. (h) HTM analysis of HUVECS treated as described in (g). Left, Cumulative frequency (C.F.) plots of the total number of WPBs as a function of the organelles’ length. N ~ 105 per condition. £, P < 10−15 for the indicated pairwise comparisons (Mann-Whitney test). Right; long WPBs (>2 μm). Bars: mean ± range for two replicate HTM experiments. (i) HUVECs treated as described and subjected to the indicated flow assay (left) to measure vWF deposition on endothelial cells (right). Quantifications were as described above. ****P < 0.0001 (Mann-Whitney test). (j) vWF deposition on HUVECs treated with vehicle (DMSO), 0.5 or 2.5 μmol/L simvastatin for 24 h and perfused as described (left). ****P < 0.0001 (Mann-Whitney test).

Fig 5: Binding of various forms of plasminogen or apo(a) to immobilized ADAMTS13, MDTCS or fibrinogen. Various forms of plasminogen (A) (Glu-Pg, Lys-Pg, mini-Pg, µ-Pg, VFK-plasmin) or apo(a) (B) at varying concentrations 0–3.5 or 0–7 µM were injected onto immobilized ADAMTS13 (A—left), MDTCS (A—right), or fibrinogen. Change in response units was monitored over time, and corrected using a blank empty flow channel. Equilibrium value response units are plotted as a function of analyte concentration. Data points are presented as mean ± standard deviation or 3 biological replicates. Data points are fitted using a one-site binding model using GraphPad Prism.