Response: Establishing the physiological significance in vivo answers the challenge of unraveling the details of GPIb-initiated signal transduction

We agree with Jackson et al that caution should be exercised in evaluating data obtained while trying to understand the details of GPIb signaling. However, we believe this caution should be extended to GPIb signaling data obtained using all in vitro signaling systems. It has not been established that GPIb-dependent calcium flux is a better indicator of physiologically significant GPIb-dependent signaling than other downstream measures such as thromboxane A2 generation, dense body secretion, or αIIbβ3-dependent platelet aggregation, nor has the precise signaling pathway prior to calcium flux been established. Nevertheless, studies from their group place src kinases and phospholipase Cγ2 upstream of calcium flux,¹  data consistent with our observations of signaling proteins upstream of thromboxane A2 generation.²  Furthermore, the widespread relationship between Bruton tyrosine kinase (Btk) and PLCγ2 signaling³  is consistent with our demonstration that Btk is required for GPIb signaling, given the clear evidence that PLCγ2 is required for GPIb signaling.^1,2 

Jackson et al caution that data obtained using suspension-based approaches, for example botrocetin (bt)/VWF stimulation of platelets, is suspect because both GPIb and αIIbβ3 signaling occur under these conditions. On the contrary, the bt/VWF system enables a clear distinction to be made between the effects of GPIb signaling versus subsequent αIIbβ3 signaling.^2,4,5  Stimulation of platelets with bt/VWF in the presence of EDTA or antibodies that prevent ligand binding to αIIbβ3, and therefore aggregation, limit the signaling response to GPIb signaling. A comparison of data obtained in the presence of EDTA or blocking antibodies with results obtained using bt/VWF-stimulated β3-deficient platelets makes this clear.⁵  Therefore, bt/VWF stimulation of platelets from KO mice can readily demonstrate if the deficient signaling molecule is required for GPIb-induced activation of αIIbβ3.^2,4,5 

Jackson et al state that caution should also be exercised in interpreting our data⁴  demonstrating that Btk is required for GPIb-dependent signaling. We used multiple approaches to make interpreting our data uncomplicated. Not only did we use anti-αIIbβ3 Ig or EDTA to prevent aggregation in the bt/VWF suspension system, we confirmed our results using platelets from mice deficient in normally functional Btk,⁴  the prediction being that if Btk is required for early events in GPIb-elicited αIIbβ3 activation, TxA2 would not be produced, and therefore αIIbβ3 would not be activated.⁵  Those are the data we obtained.⁴  Finally, we used the ferric chloride carotid artery injury system^6,7  to provide an in vivo demonstration of the physiological significance of Btk function in GPIb signaling.⁴  The results were unequivocal; Btk was required for stable thrombus formation in this GPIb-dependent system.⁴  The in vivo results confirmed the GPIb signaling data obtained using an in vitro system. Although it could be argued that the ferric chloride injury model may not reflect normal physiologic conditions, thrombus formation in that system is GPIb dependent and occurs in the absence of anticoagulants.⁴ 

Measuring calcium flux in vitro remains an important approach, but the ultimate touchstone of physiological significance is an in vivo test. Therefore in view of our in vivo data, it is clear that Btk plays an important role in GPIb-dependent signaling.