Only the strong: when antibodies hold on

Wendy E. Thomas

In this issue of Blood, Quach et al demonstrate that a rarely studied property of antibodies can be critical to immune function.1

Strong antibodies hold on to activate GPIb-IX signaling. (A) Antibodies sometimes bind to GPIbα on 2 different platelets. When the platelets move at different speeds, these GPIb-antibody complexes are pulled. (B) A strong antibody holds on tightly whereas the intraplatelet forces unfold the stalk of GPIbα. This transmits a signal inside the cell, likely by allowing the other subunits in the GPIb-IX complex to move closer together. (C) A weak antibody will let go of GPIbα before the stalk unfolds.

We usually characterize a monoclonal antibody by its epitope and affinity, but these properties have been insufficient to understand why some patients with an autoimmune disorder called immune thrombocytopenia (ITP) are refractory to standard treatments. This article points out that some immune functions or disorders require antibodies to hold on to targets that would otherwise be yanked away.

ITP is an autoimmune bleeding disorder in which antibodies bind to, and trigger clearance of, platelets. Cells are usually cleared from the bloodstream through immune mechanisms triggered by the Fc region of antibodies binding to the cell. However, some antibodies to the glycoprotein Ib-IX (GPIb-IX) complex on platelets initiate an Fc-independent clearance pathway2 by activating platelets.3 Recognition of the ligand-binding domain (LBD) of the GPIbα subunit is necessary, but not sufficient, for platelet activation. Until now, one could not predict which antibodies would activate platelets, making patients refractory to treatments targeting Fc-dependent platelet clearance. In their article, Quach et al show that antibodies must hold on to GPIbα under force to activate platelets.

Prior to this study, it was known that antibodies to the GPIbα subunit mimic the natural GPIbα ligand, von Willebrand factor (VWF), to initiate GPIbα-mediated signal transduction, platelet activation, and clearance in the liver.3 This activation pathway requires a dimeric form of the antibody, implying that GPIb-IX signals when clustered.3 However, it was soon learned that dimerization of GPIbα is not enough to activate platelets.4 Instead, GPIb-IX is a mechanoreceptor, meaning it signals when pulled.4 In the blood, fluidic shear stress causes platelets that are different distances from the vessel wall to move at different speeds, causing faster platelets to pull on slower ones when they are bound together through a dimeric molecule (see figure panel A). The GPIb-IX complex detects this pulling and signals when the stalk of GPIbα unfolds (see figure panel B).5 This explained why antibodies must target the LBD of GPIbα to activate and clear platelets, but did not explain why some antibodies target the GPIbα LBD but cannot activate platelets.3

Quach et al compare 2 antibodies, 6B4 and AK2, that both recognize the LBD of GPIbα. 6B4, but not AK2, mediated shear-dependent unfolding of the GPIbα stalk and platelet activation. The authors showed that the critical difference between the antibodies was that AK2 was simply too weak under force, letting go of GPIbα before the stalk could unfold (see figure panel C). This study therefore demonstrates that only strong antibodies can activate a mechanoreceptor. This may not be the only case where antibodies cause disease by activating mechanoreceptors because mechanosensitive proteins are common,6 so it may become necessary to develop methods to test antibody strength in clinical laboratories. Quach et al characterized the mechanical strength of the antibody-antigen interactions using methods that are not currently feasible in a clinical laboratory setting. However, they addressed the same questions for clinical samples using much more efficient assays and less expensive equipment.

This is not the first instance in which it has been discovered that the mechanical strength of recognition interactions involved in adaptive immunity can be critical to function. Natkanski et al showed that B cells generate high-affinity antibodies by pulling on and internalizing antigen clusters to discriminate between strong and weak interactions by mechanical strength.7 Liu et al showed that T cells can distinguish between agonist peptides and antagonist peptides bound to major histocompatibility complexes by the mechanical strength of the recognition interaction.8 Mechanical force is often involved in immune functions through fluidic shear stress as in the Quach study or cytoskeletal contraction as in the Natkanski and Liu studies. It remains an open question as to how often the mechanical strength of different antibody-antigen interactions may be critical in explaining the different clinical outcomes for different patients.


  • Conflict-of-interest disclosure: The author declares no competing financial interests.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.