Advertisement

Discovery and characterization of 2 novel subpopulations of aPS/PT antibodies in patients at high risk of thrombosis

Mathivanan Chinnaraj, William Planer, Vittorio Pengo and Nicola Pozzi

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Prothrombin structures. Prothrombin is a multidomain protein that adopts closed (A) and open (B) states. It comprises the N-terminal Gla-domain (blue), 2 kringles (kringle 1 [K1], red; kringle 2 [K2], green), and a serine protease domain (SP, yellow). The X-ray crystal structures of the prothrombin mutant proTCC (Protein Data Bank Identifier: 6C2W,26 left panel) and proTΔ154-167(Protein Data Bank Identifier: 5EDM,34 right panel) have been solved recently. Mutation of residues S101 and A470 to cysteine in proTCC stabilizes the closed form. The location of these 2 residues is shown in cyan. Conversely, mutation of residue Y93 to alanine in the mutant proTY93A, deletion of 14 of 26 residues of Lnk2 in the mutant proTΔ154-167, or addition of argatroban stabilizes the open form. The location of these residues is shown in magenta. The Gla-domain of both structures has been slightly modified from the original structures and adopts the calcium-bound, physiologically relevant conformation.47

  • Figure 2.

    Reactivity of aPS/PT toward closed and open conformations of prothrombin in solution. (A) IgG aPS/PT were researched in 27 APS triple-positive patients (P1-P27). Two healthy donors (HD1 and HD2), one negative control (– CTRL), and 1 positive control (+ CTRL) were included. The dashed red line (optical density at 450 nm [OD450nm] = 0.32 or 30 units) identifies the cutoff value. (B-D) Competition experiments. (B) Plasma samples (50 μL, 1:50 vol/vol) from 24 APS patients positive for aPS/PT were incubated side-by-side with a solution (50 μL, 0.72 mg/mL, 10 μM) of proTCC, proTY93A and proTΔ154-167 for 30 minutes at room temperature. The residual levels of aPS/PT were quantified by using ELISA assays. The effect of each competitor is reported as the percentage of inhibition calculated by using the following: 100 × ((OD1 – OD2)/OD1), where optical density 1 (OD1) and OD2 are the values of absorbance in the absence and presence of competitor, respectively. (C) Dose-dependent effect of proTCC (blue circles) and proTY93A (red circles) (0-15 μM) as shown for a representative patient (P7). Data were analyzed with a simple binding equation, S = S0 + (S I/IC50)/(1 + I/IC50), and the values of IC50 are reported in Table 2. (D) Plasma samples (50 μL, 1:50 vol/vol) from 24 APS patients positive for aPS/PT were incubated side-by-side with a solution (50 μL, 0.72 mg/mL, 10 μM) of proTWT in the absence or presence of 200 μM argatroban (Arg) for 30 minutes at room temperature. The residual levels of aPS/PT were quantified by using ELISA assays as described before. (E-F) Reactivity of aPT-A toward closed and open conformations of prothrombin in solution. (E) Diluted plasma samples (50 μL, 1:50 vol/vol) from 13 APS patients positive for aPT-A were incubated side-by-side with a solution (50 μL, 0.72 mg/mL, 10 μM) of proTWT, proTCC, proTY93A, or proTΔ154-167 for 30 minutes at room temperature. The residual levels of aPT-A were calculated as described above. (F) Dose-dependent effect of proTCC (blue circles) and proTY93A (red circles) (0-15 μM) as shown for a representative patient (P7). IC50 values are reported in Table 2. (G) Subpopulations of aPS/PT. Two-dimensional representation of the inhibitory effect determined in Figure 2B for proTCC (closed, x-axis) and proTY93A (open, y-axis) in the 24 APS patients positive for aPS/PT. Dashed black lines divide the plot into 4 quadrants. Group A patients (red box) are mostly located in the upper left quadrant, with a few outliers. Group B patients falls into the upper right quadrant. No aPS/PT were detected in the lower right quadrant.

  • Figure 3.

    Epitope mapping of aPS/PT antibodies from groups A and B. (A) Prothrombin fragments: GD-proT (residues 44-579); prethrombin-1, pre-1 (156-579); prethrombin-2, pre-2 (residues 285-579); fragment-1, F1 (residues 1-154); kringle-1, K1 (residues 44-155); and kringle-2, K2 (residues 156-271). Total IgG extracts from 7 patients from group A (B) (50 μL) (P4, P7, P9, P13, P14, P24, and P27) and 7 patients from group B (C) (50 μL) (P1, P3, P6, P17, P19, P21, and P25) were mixed with a solution (50 μl, 10 μM) of proTWT or specified fragment for 30 minutes at room temperature. Argatroban (Arg) was mixed with GD-proT at a concentration of 250 μM. The residual level of aPS/PT antibodies from group A (B) and group B (C) was quantified by using ELISA assays as described before. The effect of each competitor is reported as the percentage of inhibition relative to the effect of proTY93A (100%, dashed red line). (D) Visual abstract summarizing the main finding of the epitope mapping studies.

  • Figure 4.

    Effect of the membranes. Binding of proTWT (green circles), proTCC (blue circles), and proTY93A (red circles) to PS-containing liposomes monitored by surface plasmon resonance. (A) Representative sensograms for proTCC (0.125-2 μM, 1:2 dilution). (B) Plot of response units (RUs) as a function of prothrombin concentration. Solid lines were drawn according to a simple binding equation with best-fit parameters: proTWT, RU0 = 0, RU = 773 ± 14, dissociation constants (Kd) = 270 ± 15 nM; proTCC, RU0 = 0, RU = 822 ± 14, Kd = 329 ± 10 nM; proTY93A, RU0 = 0, RU = 581 ± 14, Kd = 370 ± 21nM. (C) Competitive inhibition of proTWT, proTCC, and proTY93A at 5 μM in the absence or presence of liposomes (100 μM).

  • Figure 5.

    Effect of IgG anti-prothrombin antibodies on the binding of prothrombin to phospholipids. PC:PS (75:25 wt/wt) liposomes were immobilized onto an L1 chip, and proTWT was injected at a concentration of 0.5 μM (or 0.036 mg/mL) in the absence (solid green line) or presence (0.10-0.15 mg/mL, solid red or blue line) of IgGs. Panels A-C display the effect of IgGs isolated from P4, P24, and P27 in group A (solid red line). Panels D-F display the effect of IgGs isolated from P1, P10, and P25 in group B (solid blue line).

Tables

  • Table 1.

    Laboratory characteristics of the study cohort

    Patient no.aCLAnti-β2GPIaPS/PTaPT-A
    P1++++++
    P3+++++++++++++
    P4+++++++++
    P6+++++++++-
    P7++++++++++
    P8++++++++-
    P9+++++++++++
    P10+++++++-
    P11++++-
    P12++++++-
    P13++++++++++++
    P14+++++++++++
    P15+++++++-
    P16+++++-
    P17+++++++++++-
    P18+++++++-
    P19++++++-
    P20++++++
    P21+++++++++
    P23++++++++++
    P24++++++++++
    P25++++++-
    P26++++++++++++
    P27+++++++
    • Optical density values at 450 nm for aCL, aβ2GPI, aPS/PT, and aPT-A. All patients showed lupus anticoagulant, measured as prolongation of the clotting time.3,29 P2, P5, and P22 had levels of aPS/PT below the cutoff and were therefore omitted. Scoring system for the ELISA assays: + = from 0.25 to 0.5; ++ = from 0.5 to 1.0; +++ = from 1.0 to 2.0; ++++ = from 2.0 to maximum. White = Group A; Blue = Group B.

  • Table 2.

    Values of IC50 (micromolar) for proTWT, proTCC, proTY93A, and proTΔ154-167 for IgG aPS/PT and IgG aPT-A

    VariableproTWTproTCCproTY93AproTΔ154-167Ratio (closed/open)
    aPS/PT
     P771 ± 10>804.5 ± 0.84.5 ± 0.818
     P13>80>804.1 ± 1.13.5 ± 0.819
     P2422.4 ± 525 ± 53.2 ± 0.92.8 ± 0.88
    aPT-A
     P72.6 ± 0.52.4 ± 0.52.7 ± 0.41.2 ± 0.30.9
     P133.8 ± 1.04.8 ± 1.13.7 ± 0.53.5 ± 0.51.3
     P241.7 ± 0.31.5 ± 0.52.8 ± 0.32.5 ± 0.80.5
    • Each titration was repeated twice, and the values of IC50 were determined by global fit using a simple binding equation (see "Materials and methods"). The ratio was calculated by divining the IC50 values of proTCC (closed) and proTY93A (open).