The mystery of autoantibodies solved?

In this issue of Blood, Müller-Calleja et al have identified that tissue factor is vital for the development of autoantibodies against phospholipids.¹ 

The presence of circulating antiphospholipid autoantibodies is an important risk factor for the development of thrombotic manifestations and pregnancy complications. Antiphospholipid antibody is an umbrella term for a set of different autoantibodies, including antibodies directed against the phospholipid cardiolipin and the plasma proteins β₂-glycoprotein I and prothrombin. Thanks to many studies, we more or less understand how these antibodies can induce thrombotic complications.² Guidelines for the treatment of the clinical manifestations have been developed.³ However, why some individuals develop antiphospholipid antibodies is largely unknown. The classic theory for the origin of autoantibodies is molecular mimicry: the structural similarity between foreign and self-proteins of human individuals. Infections with many different viruses and bacteria have been associated with the development of antiphospholipid antibodies.⁴ These infection-induced antiphospholipid antibodies are mostly transient and not associated with clinical features of the antiphospholipid syndrome. Considering the frequency and ease that antiphospholipid autoantibodies are formed after infections in response to a large number of dissimilar microorganisms, there must be a more basic answer for the frequent presence of these autoantibodies in the circulation.⁴ In some individuals, these antibodies become persistent and associated with thromboembolic and pregnancy complications. Why only in some individuals these autoantibodies switch from transient to permanent autoantibodies is completely unknown. The general consensus is that these infection-related antiphospholipid antibodies are distinct from the pathogenic persistent antibodies, although there is no scientific evidence supporting this distinction. Interestingly, pathogenic anti–β₂-glycoprotein I antibodies are directed against an epitope in its first domain and are cryptic in the mature protein. Immunization of mice with a fragment of β₂-glycoprotein I is enough to induce the autoantibodies.⁵ A conformational change of the target protein is essential for recognition by the autoantibodies.

Müller-Calleja et al show that tissue factor is not only an essential player in the autoantibody-induced thrombotic complications, it is also crucial for the development of autoreactive B1 cells that generate the antiphospholipid antibodies. Inhibition of tissue factor signaling with the nematode anticoagulant protein NAPc2 inhibits the development of antibodies against β₂-glycoprotein I and endothelial cell protein C receptor, the protein that mediates the binding of antibodies to cardiolipin. To our knowledge, this is the first publication showing that tissue factor is an essential intermediate for the generation of antiphospholipid antibodies. The most important question that remains is whether these tissue factor–mediated autoantibodies are the transient antiphospholipid antibodies, as often observed after infectious diseases, or whether these autoantibodies are the persistent antibodies responsible for the thrombotic complications. The authors could easily test this in their mouse models. The observation that infection with single-stranded RNA severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) also prevented antibody production in mice lacking the cytoplasmatic domain of tissue factor suggests that at least the formation of transient autoantibodies is tissue factor dependent. The antiphospholipid antibodies in COVID-19–infected patients are usually transient, and most studies examining clinical outcomes in patients with COVID-19 with and without antiphospholipid antibodies did not find a correlation between antibodies and thrombotic complications.^6,7 Moreover, the anti–β₂-glycoprotein I autoantibodies found in patients infected with COVID-19 are not directed against the first domain of β₂-glycoprotein I.⁷ Many studies have shown that autoantibodies against domain I of β₂-glycoprotein I are the only antibodies responsible for the clinical manifestations that characterize the antiphospholipid syndrome (see figure).⁸ 

Are transient autoantibodies an early phase in the development of persistent autoantibodies and, if so, what is the reason for the switch? There are no indications that there is a genetic predisposition, as the risk of developing the antiphospholipid syndrome is not increased within families. Some publications, including a recent publication from the authors of the Blood article discussed here, suggest that that antiphospholipid antibodies, including anti–β₂-glycoprotein I antibodies, belong to the natural occurring antibody repertoire.⁹ Natural antibodies are defined as antibodies that are present in the circulation before an antigen stimulus. Natural antibodies have been described as polyreactive (ie, they can react with several unrelated antigens, including self-antigens). Natural antibodies are secreted by B1 cells, the same cells found to be active in the current study. Antigen-driven maturation in combination with epitope shuffling can result in a shift from transient to persistent. It is possible that the antigens recognized by antiphospholipid antibodies react to different infections and stimulate an antigen-driven change in antiphospholipid antibody expression, the latter process dependent on the presence of tissue factor.

Conflict-of-interest disclosure: P.G.d.G. declares no competing financial interests.