Platelets contain 40–100 times as much transforming growth factor β1 (TGF-β1) as other cells, and release it as an inactive (latent) complex. Although several TGF-β1 activating factors have been identified in various cell types, it is still not known whether and how platelet TGF-β1 becomes activated in vivo. In the present study, cell-free human washed platelet releasates (supernatant after stimulation with thrombin 0.125U/ml for 5 min) or serum samples were subjected to either stirring (1,200 rpm) in an aggregometer or a shear stress of 1800 s−1 in a cone and plate device at 37°C. Activation of TGF-β1 was analyzed by ELISA and selectively confirmed using a cell-based plasminogen activator inhibitor 1 (PAI-1) luciferase reporter assay. TGF-β1 was maximally released within 5 min after stimulation of platelets with thrombin (80 ± 23 ng/ml; n=3), but only 0.2% was active. TGF-β1 activity increased slowly and progressively under shear, reaching a maximum of ∼5% of total TGF-β1 (4.1 ± 1.5 ng/ml) after 1 h. In sharp contrast, in two experiments conducted without shear, active TGF-β1 constituted 0.04 and 0.2% of total at 0 time and remained nearly the same (0.1 and 0.2% of total) after 2 h. Activation of TGF-β1 in serum under shear occurred more slowly, but also reached a maximum of ∼5% of total at 2 h (4.1 ± 1.3 ng/ml active; 76 ± 14 ng/ml total; n=6). Without shear at 0 time, active TGF-β1 constituted 0.02 and 0.06% of total and at 2 h it was still only 0.05 and 0.07% of total. To asses whether TGF-β1 becomes activated in vivo during thrombosis, we induced thrombi in the carotid arteries of C57/BL mice with ferric chloride (8%) for 3 min and then removed the platelet-rich thrombi that formed after 5 or 120 min. The arteries were excised (∼4 mm) and the thrombi were removed and dispersed in buffer (200 μl) on ice for 1 h. Total TGF-β1 recovered from 5 min thrombi was 3.4 and 4.8 ng/ml and total TGF-β1 recovered from 120 min thrombi was 3.7 and 0.7 ng/ml. Active TGF-β1 could be detected in thrombi after 5 min and constituted 1.7 ± 1.1% of the total TGF-β1 recovered (n=3). After 120 min, active TGF-β1 could also be detected in thrombi and it constituted 5.1 ± 3.0% of the total recovered (p=0.14 compared to 5 min value; n=3). In sharp contrast, only 0.05% of TGF-β1 released by thrombin from mouse platelets in vitro was active without stirring or shear (n=3). These data indicate that shear can dramatically enhance TGF-β1 activation after release from platelets. Moreover, TGF-β1 activation occurs in vivo during platelet-rich thrombus formation and thus platelets may be an important source of active TGF-β1 in cardiovascular disease and wound healing.

Author notes

Disclosure: No relevant conflicts of interest to declare.