
Blood, 15 May 2002, Vol. 99, No. 10, pp. 3491-3491
Does VWF bind collagen after all?
Platelets adhere to sites of vascular injury to form an initial
platelet plug, and this process requires von Willebrand factor (VWF) to
bridge platelet membrane glycoprotein Ib (GPIb) and ligands in
exposed connective tissue. The GPIb binding site is known to reside in
VWF domain A1, which is 1 of 3 tandemly repeated 24 kd A domains in the
VWF subunit. But the nature of the VWF-connective tissue interaction
remains controversial. VWF binds several types of fibrillar collagen in
vitro, and the major collagen binding site is within VWF domain A3. A
second site may be present in domain A1, but domain A1 cannot sustain
normal platelet adhesion to collagen if domain A3 is deleted. Sites
other than A3 may bind connective tissue, and connective tissue ligands
other than fibrillar collagens may bind VWF. For example, nonfibrillar
collagen type VI binds domain A1 rather than domain A3, and monoclonal
antibodies have been prepared that prevent VWF binding to fibrillar
collagen or subendothelial connective tissue, but not both. Therefore, the contribution of VWF-collagen binding to the initial stages of
platelet plug formation remains poorly defined.
Wu and colleagues (page 3623) now have shown that a monoclonal
antibody
to VWF domain A3 that inhibits VWF-collagen binding in
vitro can prevent
platelet-rich thrombi from forming in damaged
femoral arteries in vivo. At higher doses, the antibody also prolongs
the skin bleeding time of baboons. These findings demonstrate the
physiologic importance of VWF domain A3 in platelet adhesion and
suggest that the relevant connective tissue ligand is a fibrillar
collagen. The data do not exclude a role for other VWF binding sites or
other connective tissue ligands, but they do suggest that VWF sites
other than domain A3 may be insignificant for connective tissue
interactions. Furthermore, the antithrombotic effect of the anti-VWF
antibody persisted for several hours after a single injection, raising
the possibility that interference with the VWF domain A3-collagen
interaction could be a feasible antithrombotic strategy.
Evan Sadler
Howard Hughes Medical Institute, Washington University School of
Medicine