Annexin A2: better left alone

Alisa S. Wolberg and Robert A. S. Roubey

Comment on Zhang and McCrae, page 1964

Zhang and McCrae demonstrate that APLA/b2GPI-mediated endothelial cell activation occurs via dimerization of annexin A2 molecules on the cell surface.

The association of antiphospholipid antibodies (APLAs) with thrombosis is well established, and there is growing evidence that APLAs themselves contribute to hypercoagulability. The mechanisms of APLA-associated hypercoagulability are less clear, however. A confusing variety of antibody effects has been proposed. One important line of research involves the activation of endothelial cells by certain APLAs, specifically those directed against β2-glycoprotein I(β2GPI). In their current study, Zhang and McCrae make an important next step in this endeavor.

Previously, APLAs, in a β2GPI-dependent manner, were shown to stimulate endothelial cells to a procoagulant phenotype by increasing monocyte adhesion and the expression of E-selectin, vascular cell adhesion molecule-1, and intracellular adhesion molecule-1.1,2 These events presumably require the transduction of a signal into the cell and, therefore, the involvement of a cell surface receptor. Since the effects of APLA/β2GPI do not appear to be mediated by Fc receptors,1 investigators posited the existence of a cell surface receptor for β2GPI. In previous work, McCrae and colleagues identified annexin A2 as a high-affinity receptor for β2GPI on endothelial cells (Ma et al3). The current paper ties this finding to the earlier functional studies, demonstrating that cross-linking of annexin A2 by APLA/ β2GPI activates endothelial cells (see figure).


Model for annexin A2–dependent endothelial cell activation induced by β2GPI and anti-β2GPI antibodies, or anti–annexin A2. See the complete figure in the article beginning on page 1964.

The key observation by Zhang and McCrae is that cross-linking of annexin A2 on the endothelial cell surface induces the expression of cell adhesion molecules. Cross-linking was achieved using bivalent anti–annexin A2 antibodies or APLA/β2GPI. The critical role of bivalent cross-linking was demonstrated using monovalent antibody fragments that (a) did not activate cells, and (b) blocked the effects of bivalent antibodies.

These findings establish at least 2 important points. First, a cell surface receptor (ie, annexin A2) is critical in mediating the in vitro effects of APLAs on endothelial cells. Second, the data highlight the role of immunoglobulin G (IgG) bivalency. Antibody bivalency has been shown to explain how antibodies to β2GPI, a relatively weak phospholipid-binding plasma protein, lead to high avidity binding of IgG-β2GPI complexes to phospholipid membranes. Now, bivalency has been shown to play a critical role in cross-linking annexin A2 and transducing an activation signal to cells.

As noted by the authors, important questions remain. How does annexin A2, which does not have a transmembrane domain, transduce a signal? Recently, Raschi et al4 have demonstrated that APLA activation of endothelial cells occurs via the MyD88 pathway that is commonly associated with toll-like receptor (TLR) signaling. Does annexin A2 signal through the MyD88 pathway? Is annexin A2 physically associated with a TLR? In addition to transducing a signal, what is the effect of APLA/β2GPI on annexin A2's role as a coreceptor for plasminogen and tissue plasminogen activator? Does binding of APLA/ β2GPI to annexin A2 inhibit plasmin generation? Once annexin A2 is cross-linked by APLA/β2GPI, what is the fate of this complex? If it is internalized or shed, the amount of annexin A2 available to support plasmin generation may be decreased. Finally, and perhaps most importantly, might the APLA/β2GPI/annexin A2 complex provide a molecular target for antithrombotic therapies in the antiphospholipid syndrome? ▪