Abstract
HIT is an iatrogenic complication of heparin therapy caused by antibodies that recognize the platelet chemokine, platelet factor 4 (PF4), complexed to heparin or to cellular glycosaminoglycans (GAG). Unlike most other immune thrombocytopenias, HIT is markedly prothrombotic. We have proposed that this prothrombotic tendency is due to binding of pathogenic antibodies to PF4 complexes attached to the surface GAGs expressed by all intravascular cells, including platelets with their relatively low affinity surface GAGs, chondroitin sulfates, and monocytes with their higher affinity membrane GAGs, heparan and dermatan sulfates. Using isolated monocytes from healthy volunteers, we show by scanning electron microscopy that the addition of 10-50 µg/ml of recombinant human PF4 causes the appearance of ∼200 nm “knobs” on the cell surface. Subsequent addition of a HIT-like monoclonal antibody KKO at 50 µg/ml to the PF4-coated cells markedly alters their surface with the appearance of larger, up to 1-2 µm, membrane “blebs”. These blebs increase in size over time (15-60 minutes) and are shed from the cells. After shedding of these blebs, the monocytes lose their typical ruffled surface and appear spherical. These surface changes in the presence of KKO and PF4 are not seen in the presence of PF4 and 50 µg/ml of the anti-PF4 monoclonal antibody RTO, which does not induce the prothrombotic state of HIT. Platelets in suspension exposed to PF4 and KKO show by scanning electron microscopy similar knobs on their surface, but only minimally form blebs or microvesiculate. Platelets spread on fibrinogen in culture medium stimulated with PF4 and KKO and observed by hopping probe ion conductance microscopy progressively developed surface protrusions over a period of an hour, becoming more spherical. This morphological change was also observed in platelets exposed to IgG purified from 5 patients with HIT, but not when the KKO Fab fragment was tested. Neither PF4 alone nor with RTO antibody induced this morphological transition. Exposure to KKO plus PF4 for an hour induced minimal microvesiculation of platelets as measured by flow cytometry. Platelets adherent to fibrinogen underwent a similar morphological transition and did not microvesiculate after adding 50 mM of a thrombin-receptor activating peptide, whereas ADP-stimulated platelets rapidly microvesiculated during the same period of time. We believe that the “knobs” observed on monocytes and platelets represent aggregates of PF4-GAG complexes that are the targets of HIT antibodies. Bleb formation on monocytes and morphological transition of platelets result from clustering of knobs caused by bivalent HIT antibodies which cross-link Fc receptors. These blebs are released from monocytes, potentially becoming the microparticles found in the plasma of patients with HIT. In contrast, platelets treated with KKO plus PF4 showed minimal microvesiculation. This finding differs from reported high platelet microparticle counts in the plasma of HIT patients, suggesting that additional factors may be required to induce platelets to microvesiculate. Our images represent the first visualization of surface events when platelets and monocytes assume an active prothrombotic state. Whether these are unique to HIT or have wider applicability to the changes that occur in other prothrombotic, proinflammatory states needs to be addressed.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.