Fibrinogen is necessary for platelet function in vivo after all

Ligand binding to the integrin αIIbβ3 (glycoprotein IIb/IIIa [GPIIb/IIIa]) is required for the formation of hemostatic platelet plugs and platelet-based arterial thrombi. A number of macromolecular ligands bind to the activated conformation of αIIbβ3. The ligand necessary for platelet aggregation in vitro is fibrinogen, but the situation in vivo appears to be substantially more complex. For example, fibrinogen-null mice do surprisingly well. Despite a 30% incidence of overt bleeding at birth, a predisposition to fatal spontaneous abdominal hemorrhage as juveniles and adults, uniformly fatal bleeding early in pregnancy, and absence of in vitro platelet aggregation, more than 90% survive the neonatal period, and, depending on their genetic background, 60% to 90% survive to at least 70 days of age.¹  Moreover, mice whose fibrinogen lacks the 5 carboxyl-terminal γ chain amino acids (γΔ5 fibrinogen) and, consequently, cannot bind to αIIbβ3 and support platelet aggregation, have an even more benign phenotype, although they are unable to control blood loss following a surgical challenge and occasionally develop fatal neonatal bleeding events.² FIG1 

von Willebrand factor (VWF) is a second αIIbβ3 ligand. Although VWF deficiency in both humans and mice is responsible for the hemostatic disorder von Willebrand disease (VWD), bleeding in VWD is generally mild and fibrinogen-dependent platelet aggregation is preserved. Furthermore, VWF-null mice manifest a relatively normal phenotype with only a modest susceptibility for spontaneous bleeding, despite a markedly prolonged bleeding time and absence of thrombi when mesenteric arterioles are injured by ferric chloride.³ 

Because fibrinogen and VWF are thought to be the predominant αIIbβ3 ligands, it came as another surprise when Ni et al⁴  reported that 77% of mice who were null for both fibrinogen and VWF survived to adulthood. Moreover, platelet thrombi were inducible in the ferric chloride injury model, although their appearance was delayed and they were unstable as manifested by frequent downstream embolization. These observations suggest that another αIIbβ3 ligand, likely fibronectin, can participate in formation of platelet aggregates, particularly when the concentrations of fibrinogen and VWF are limiting.

In the current issue of Blood, Ni and colleagues (page 3609) provide additional insight into the mechanism of platelet thrombus formation in vivo. Because fibrinogen-null mice have a more severe bleeding phenotype than γΔ5-fibrinogen mice, one would suppose that fibrin formation is essential to form a stable platelet thrombus. To test this supposition, Ni and colleagues observed ferric chloride-induced thrombus growth in arterioles of γΔ5-fibrinogen mice using intravital microscopy. Because γΔ5-fibrinogen clots normally, this mutant can distinguish fibrinogen's ability to support platelet aggregation from the effects of fibrin polymer formation. They found no difference in initial platelet adherence between γΔ5-fibrinogen and wild-type mice, reinforcing the role of VWF in mediating platelet adhesion at high shear. On the other hand, whereas subsequent thrombus growth was paradoxically faster in the γΔ5-fibrinogen mice, there was also persistent shedding of large numbers of emboli before vessel occlusion occurred, implying that fibrinogen bound to αIIbβ3 was necessary to form stable thrombi at the shear rates present in arterioles. Interestingly, they also observed an increase in the amount of fibronectin in the α granules of platelets from γΔ5-fibrinogen mice, suggesting that there is normally competition between fibrinogen and fibronectin for the αIIbβ3-mediated uptake of these proteins.

The experiments performed by Ni and colleagues reveal that fibrinogen binding to αIIbβ3 plays an essential, and unexpected, role in thrombus stability, likely by strengthening the adhesion of platelets as they are incorporated into a growing thrombus. In vitro experiments have necessarily focused on individual αIIbβ3 ligands. The lesson to be learned from the current study by Ni and colleagues, and from the others cited above, is that, in vivo, nature has made use of the entire repertoire of αIIbβ3 ligands to ensure stable and hemostatically effective platelet-mediated hemostasis.