Platelet dense granule secretion defects may obscure α-granule secretion mechanisms: evidence from Munc13-4–deficient platelets

To the editor:

We were very interested to read in Blood the recent papers by Meng et al, “Defective release of α granule and lysosome content from platelets in mouse Hermansky-Pudlak syndrome models” and by Sharda et al, “Defective PDI release from platelets and endothelial cells impairs thrombus formation in Hermansky-Pudlak syndrome.”^1,2  In these papers, the authors show that platelet adenosine diphosphate (ADP) secretion from dense granules is an important autocrine regulator of α-granule secretion. We have reached essentially the same conclusion, however, from a different perspective. We believe that, together, our data show that experimental manipulations or conditions that disrupt dense granule secretion may indirectly affect α-granule secretion and should be taken into account when investigating the mechanisms of α-granule secretion.

We are interested in the molecular machinery that controls platelet granule secretion. The SNARE regulator, Munc13-4, was shown to be an essential factor for dense granule secretion.³  Consistent with this, Unc13d^Jinx mice (which lack Munc13-4) have prolonged tail bleeding and are protected from arterial thrombosis and cerebral infarct progression.^3-5  It has also been previously shown that Unc13d^Jinx platelets have a substantial, but incomplete, suppression of α-granule secretion,³  an effect that we also observe. Washed platelets were stimulated with PAR4-activating peptide (PAR4-AP) (300 µM) and surface expression of CD62P was used as a marker of α-granule secretion. PAR4-AP–induced CD62P expression was significantly less in Unc13d^Jinx platelets compared with wild-type (WT) platelets (Figure 1A). This suggests that almost all α-granule secretion is dependent on Munc13-4. However, the block to α-granule secretion in the absence of functional Munc13-4 was substantially rescued by the addition of ADP (Figure 1A-B). Therefore, there must also be a Munc13-4–independent pathway that is capable of inducing substantial α-granule secretion.

ADP alone was not able to induce detectable CD62P expression under these conditions, suggesting that ADP acts synergistically to enhance PAR4-AP–induced α-granule secretion. In the presence of the P2Y₁₂ antagonist AR-C69931MX, there was a marked suppression of CD62P surface expression in both WT and Unc13d^Jinx platelets, and the ADP-mediated rescue of secretion was ablated (Figure 1B). The P2Y₁ antagonist, MRS2279, had no effect. In addition, we found that 5-HT was not able to induce α-granule secretion, and neither did it synergise with PAR4-AP to rescue the response in the Unc13d^Jinx platelets (data not shown), suggesting that 5-HT has no role in the regulation of secretion under these conditions. Therefore, secreted ADP, acting through the P2Y₁₂ receptor, synergizes with PAR4-AP to induce Munc13-4–independent α-granule secretion, but the presence of this pathway is obscured by the complete loss of dense granule secretion in Unc13d^Jinx platelets.

To confirm that secreted ADP regulates PAR4-AP–induced α-granule secretion, we also investigated the role of ADP in the release of the α-granule cargo, PF4 (Figure 1C). PAR4-AP–stimulated Unc13d^Jinx platelets released significantly less PF4 than WT platelets. Co-stimulation with ADP induced more PF4 release in Unc13d^Jinx platelets than PAR4-AP alone (34.3 ± 2.2% of total PF4 with PAR4-AP+ADP compared with 19.0 ± 2.3% with PAR4-AP alone), again suggesting that loss of secreted ADP contributes to the α-granule secretion defect in these platelets.

These data demonstrate that the disruption of dense granule secretion can alter the apparent contribution of different proteins to α-granule secretion. For example, under our stimulation conditions, we would have concluded that Munc13-4 is a major regulator of PF4 release, whereas it may be released in a largely Munc13-4–independent manner that is enhanced by secreted ADP. There was still a small, statistically significant, difference between PF4 release from WT and Unc13d^Jinx platelets in the presence of ADP, suggesting that there may be a small direct contribution of Munc13-4 under these conditions. Alternatively, this may reflect a small contribution of other dense granule components, or that exogenous ADP does not fully mimic locally high concentrations of secreted ADP. The relative contribution of the Munc13-4–dependent pathway and the Munc13-4–independent pathway may depend on the precise conditions of platelet stimulation. However, we believe that our data raise the more general question: How much do we know about the molecular mechanisms of α-granule release in platelets? It will be particularly interesting to re-evaluate knockout mice studies that show defects in dense and α-granule secretion, to determine which proteins regulate α-granule secretion directly, and which act indirectly through dense granule secretion.