Background/Purpose:

Circulating IgG antiphospholipid antibodies (aPL) against β2-glycoprotein I (aβ2GPI) are a serological hallmark for diagnosis of the antiphospholipid syndrome (APS). We and other groups have shown that aPL targeting domain I (DI) of β2GPI (aDI) are APS-specific, predominantly correlating with (venous) thrombosis. We have also demonstrated that recombinant DI inhibits aPL-induced thrombosis in a mouse microcirculation model. To date however, no study has confirmed a direct pathogenic link between aDI and features of the APS. We have now employed the same mouse model to determine the thrombogenic potential of affinity purified polyclonal aDI IgG derived from APS sera.

Methods:

Serum from one female APS patient was incubated with his-tagged DI coupled to nickel beads to adsorb aDI antibodies. The bead-serum mix was then spun, serum re-collected and antibodies bound to DI-coupled beads were eluted. IgG from re-collected serum (aDI-poor) and eluted fractions (aDI-rich) was then obtained by protein G purification. Serum and IgG fractions (100mg/mL) were tested for aCL (GPLU), aβ2GPI (GBU, in-house calibrator) and aDI (GDIU, in-house calibrator) activity.

For in vivo experiments, CD1 mice weighing between 32-39g were injected twice with 100µg/mL aDI-poor or aDI-rich IgG, or IgG from healthy volunteers (NHS-IgG) as a control (4-5 animals per group). 72h after the first injection, the size (µm²) of induced thrombi in the femoral vein was determined (Circulation 1996;94:1746-1751). Tissue factor (TF) activity (pM/mg.mL^-1protein) was determined in homogenates of pooled carotid arteries and peritoneal macrophages using a chromogenic assay. Mouse serum was obtained on the day of surgery and tested for the presence of circulating whole human IgG.

Results:

Purified aDI-rich IgG displayed high aCL (90GPLU), aβ2GPI (95GBU) and aDI (50GDIU) activity whilst aDI-poor IgG displayed high aCL (90GPLU) but reduced aβ2GPI (47GBU) and aDI (17GDIU) activity. NHS-IgG was negative in all assays. aDI-rich IgG induced significantly larger thrombi compared to aDI-poor and NHS-IgG (p<0.001). In addition, aDI-rich IgG induced the greatest increase in TF activity in carotids (1.4 fold) and peritoneal macrophages (3.3 fold) compared to NHS-IgG. In contrast, aDI-poor IgG induced smaller thrombi and less macrophage TF activity compared to aDI-rich IgG and did not increase carotid TF activity above that of NHS-IgG (Table 1).

Table 1. Effects of polyclonal IgG in an in vivo model of thrombosis
Mice/ treatment	Thrombus size (µm2)	TF activity (†) carotids (pM/mg.mL-1 protein)	TF activity (†) peritoneal macrophages (pM/mg.mL-1 protein)
NHS-IgG	525 ± 136	149.8 ± 21.1	30.6 ± 4.1
aDI-poor	953 ± 258 *	134.9 ± 29.1	55.3 ± 10.1
aDI-rich	2076 ± 511 **	212.9 ± 49.7	101.2 ± 14.1
(*) Statistically significant <0.001 to NHS-IgG (**) Statistically significant <0.001 to NHS-IgG and aDI-poor (†) mean of two measurements

Conclusion:

This is the first study to directly demonstrate the thrombogenic potential of affinity purified aDI IgG in vivo. Despite aDI-poor IgG retaining aCL and, to a lesser extent, aβ2GPI activity, significantly larger thrombi and elevated TF activity were induced with aDI-rich IgG. Our findings support the concept that although circulating aPL recognizing different domains of b2GPI can be pathogenic, the major population that drive thrombosis are directed against DI.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/affinity-purified-antibodies-directed-to-domain-i-of-%ce%b22gpi-are-pathogenic-in-a-mouse-model-of-thrombosis/