To prevent excessive blood loss at sites of trauma and yet avoid life-threatening thrombosis from vessel occlusion, a finely tuned hemostatic balance must be maintained. A highly complex set of interactions maintains this hemostatic balance, and elucidation of the intricacies of hemostasis has been a major challenge in our field. Multiple approaches, including in vitro, ex vivo, and in vivo analyses, have been deployed in our efforts to delineate the mechanisms that regulate thrombus formation. For more than a decade, transgenic mice have provided an incisive means to dissect the complexities of the hemostatic system. More recently, sophisticated approaches for thrombus imaging have been applied to visualize and quantify the processes of thrombus formation and dissolution. Some of these analyses have supported the prevailing views of these processes. Such results provide assurances that existing tenets have biological validity and that the murine models are translatable to human physiology. As an example that is pertinent to the article discussed below, a knockout of the integrin β3 subunit induces a bleeding syndrome that recapitulates Glanzmann's thrombasthenia [1]. These data are compatible with extensive in vitro observations documenting the importance of αIIbβ3 as a platelet receptor for adhesive proteins that mediate platelet aggregation and thrombus development [2]. The results obtained from interrogation of other transgenic mouse models have often been unexpected. Despite a plethora of in vitro and ex vivo studies that document the predominant roles of von Willebrand factor (VWF) and fibrinogen as the major ligands for αIIbβ3 in mediating platelet aggregation, mice deficient in both VWF and fibrinogen still form a thrombus [3]. Together, these observations suggest that other ligands of αIIbβ3 can also regulate the participation of platelets in thrombus formation.

Although vitronectin is a known ligand of αIIbβ3 [4,5], its binding to the receptor was not viewed as being functionally unique. It was simply one of many proteins that contains an RGD sequence, and many proteins can bind to αIIbβ3, as well as to other integrins, through their RGD sequences [6]. This mindset has tended to diffuse interest in vitronectin as a modulator of platelet responses and has turned the spotlight on various other ligands of αIIbβ3. In this issue of the Journal of Thrombosis and Haemostasis, Reheman et al. [7] have revisited the role of vitronectin in thrombus formation and vessel occlusion. Their study makes extensive use of the vitronectin-deficient (Vn−/−) mouse and uses intravital microscopy to evaluate thrombus formation and stability. The Vn−/− mice had been developed and previously characterized [8], and no major spontaneous phenotype had been identified. It is, in fact, the absence of an overt phenotype in Vn−/− mice that permits assessment of their response to a thrombotic challenge without other confounding complications. Previous studies of thrombosis in the Vn−/− mice had yielded a mixed picture with evidence that vitronectin could either support or inhibit a thrombotic response [9]. Consequently, the role of vitronectin in thrombus remained out of the spotlight.

In the Reheman study, two different models of arteriole thrombosis were compared in the Vn−/− and wild-type (WT) mice: (i) a mesentery thrombosis model in which injury was induced by FeCl3, and (ii) a cremaster muscle model in which injury was induced with a laser. The use of two different models employing two modes of injury and at two different anatomic sites adds a degree of elegance to the study, particularly as the two injury models yielded similar results. Thrombus formation was monitored by intravital microscopy using fluorescently labeled platelets or antibodies to platelets or to fibrin for tracking and quantification. In the mesenteric arteriole thrombosis model, initial platelet deposition on the injured vessel wall was not different in Vn−/− and WT mice. Formation of the initial thrombus was also not significantly different in the two genotypes. However, parameters of thrombus stability were significantly perturbed by the absence of vitronectin. This instability was reflected in a substantial increase in the number of emboli that formed either prior to or after vessel occlusion. As a result of this instability, it took longer for vessels to occlude and they reopened with greater frequency in the Vn−/− mice. In the cremaster arteriole thrombosis model, the rates of thrombus formation after injury, whether monitored in terms of platelet or fibrin deposition, were similar in Vn−/− and WT mice. However, less fibrin and platelets accumulated in the thrombi in the Vn−/− mice over time and their amounts fluctuated substantially. These findings are consistent with clot instability due to/leading to embolization as was observed in the mesentery thrombosis model. Thus, both models point to a vital role of vitronectin in clot stabilization and to the fact that its absence leads to frequent embolization.

Vitronectin can be derived from multiple origins. It is a major plasma protein, circulating in the blood at 200–500 μg mL−1. It can also be present in extracellular matrices and can support adhesion of platelets as well as other cells expressing integrin αVβ3, originally known as the vitronectin receptor. In addition, vitronectin is present in platelet α-granules [10] and is secreted upon stimulation of the cells. VWF, clearly a critical player in thrombus formation, can be derived from these same three compartments (plasma, vessel wall, and platelets). By virtue of its broad and abundant distribution, there is ample opportunity for vitronectin to influence platelet responses. At the same time, its effects may be complicated as some of these influences may be counterbalancing. Adding to the complexity, vitronectin can exist in various ‘activation’ and aggregation states [11,12], which can have distinct functional activities.

To identify ways in which vitronectin derived from the various compartments might influence thrombus stability, Reheman et al. performed aggregation studies using platelets or plasma from Vn−/− or WT mice. Using ADP as an agonist, they showed that plasma vitronectin is an inhibitor of platelet aggregation. Platelet aggregation in response to this agonist was enhanced in Vn−/− platelet-rich plasma (PRP) compared with WT PRP. When Vn−/− platelets were placed in WT plasma, aggregation was reduced compared with that observed in Vn−/− plasma, and when WT platelets were placed in Vn−/− plasma, aggregation was enhanced relative to that observed in WT plasma. Finally, addition of vitronectin to Vn−/− PRP suppressed aggregation of both Vn−/− platelets and WT platelets. All data consistently point to a suppressive effect of vitronectin on platelet aggregation. This phenomenon was also observed when platelet aggregation was induced at certain shear rates in a plate-and-cone viscometer. Aggregation was more extensive in Vn−/− than in WT blood. Such an inhibitory effect would be anticipated if plasma vitronectin competed with fibrinogen for binding to αIIbβ3 but binding of vitronectin to the integrin does not support platelet aggregation. Unfortunately, this inhibitory effect does not account for the suppression of thrombus stability in Vn−/− mice; that is, thrombus formation should be enhanced in animals lacking vitronectin, but it was not.

The authors then examined the influence of platelet vitronectin in aggregation. Aggregation of washed platelets in response to thrombin, which would induce vitronectin release of α-granules, was assessed. The aggregation of Vn−/− platelets was substantially reduced in comparison with WT platelets. Under conditions where WT platelets exhibited both first-wave and second-wave aggregation curves, Vn−/− platelets showed no second-wave aggregation. Microscopically, Vn−/− platelet aggregates were smaller than WT aggregates. Thus, it is this anti-aggregatory activity of platelet vitronectin that is consistent with the instability of clots formed in Vn−/− mice.

The balance between the pro-aggregatory and anti-aggregatory activities of vitronectin is likely to be very delicate. The nature of the thrombotic stimulus and other modulators may determine which of these effects of plasma and platelet vitronectin are more influential. Moreover, other functions of vitronectin may also influence thrombus formation and stability. Hence, it is not surprising that independent studies of the role of vitronectin in thrombosis have not always yielded consistent results. The study of Reheman et al. [7] establishes a clear contribution of vitronectin to thrombus stability in vivo and suggests a role for platelet vitronectin in this phenomenon. Hence, vitronectin enters back into the spotlight as a modulator of thrombus formation and vascular occlusion.


  1. Top of page
  2. Acknowledgements
  3. References

I would like to thank Olga Stenina, PhD, for her expertise in reviewing the subject of this Commentary.


  1. Top of page
  2. Acknowledgements
  3. References
  • 1
    Hynes RO, Hodivala-Dilke KM. Insights and questions arising from studies of a mouse model of Glanzmann thrombasthenia. Thromb Haemost 1999; 82: 4815.
  • 2
    Quinn MJ, Byzova TV, Qin J, Topol EJ, Plow EF. Integrin αIIbβ3 and its antagonism. Arterioscler Thromb Vasc Biol 2003; 23: 94552.
  • 3
    Ni H, Denis CV, Subbarao S, Degen JL, Sato TN, Hynes RO, Wagner DD. Persistence of platelet thrombus formation in arterioles of mice lacking both Willebrand factor and fibrinogen. J Clin Invest 2000; 106: 38592.
  • 4
    Asch E, Podack ER. Vitronectin binds to activated platelets and plays a role in platelet aggregation. J Clin Invest 1990; 85: 13728.
  • 5
    Thiagarajan P, Kelley KL. Exposure of binding sites for vitronectin on platelets following stimulation. J Biol Chem 1988; 263: 30358.
  • 6
    Ruoslahti E. RGD and other recognition sequences for integrins. Annu Rev Cell Biol 1996; 12: 697715.
  • 7
    Reheman A, Gross P, Yang H, Chen P, Allen D, Leytin V, Freedman J, Ni H. Vitronectin stabilizes thrombi and vessel occlusion but plays a dual role in platelet aggregation. J Thromb Haemost 2005; 3: 87583.
  • 8
    Zheng X, Saunders TL, Camper SA, Samuelson LC, Ginsburg D. Vitronectin is not essential for normal mammalian development and fertility. Proc Natl Acad Sci USA 1995; 92: 1242630.
  • 9
    Fay WP, Parker AC, Ansari MN, Zheng X, Ginsburg D. Vitronectin inhibits the thrombotic response to arterial injury in mice. Blood 1999; 93: 182530.
  • 10
    Seiffert D, Schleef RR. Two functionally distinct pools of vitronectin (Vn) in the blood circulation: identification of a heparin-binding competent population of Vn within platelet alpha-granules. Blood 1996; 88: 55260.
  • 11
    Izumi M, Yamada KM, Hayashi M. Vitronectin exists in two structurally and functionally distinct forms in human plasma. Biochim Biophys Acta 1989; 990: 1018.
  • 12
    Stockmann A, Hess S, Declerck P, Timpl R, Preissner KT. Multimeric vitronectin. Identification and characterization of conformation-dependent self-association of the adhesive protein. J Biol Chem 1993; 268: 2287482.