Is NADPH oxidase the target of phosphoinositide-3-kinase-mediated ROS generation in platelets?

To the Editor

We read with great interest the recent paper by Liu and colleagues on the role of phosphoinositide 3-kinase (PI3K) in platelet activation and thrombosis.¹ Liu et al. demonstrated that mice lacking platelet-specific VPS34, a PI3K isoform, exhibit diminished platelet activation and aggregation and are protected from accelerated thrombosis. This phenotype was associated with loss of translocation of the p40phox cytoplasmic subunit of NADPH oxidase to the plasma membrane, suggesting that VPS34 may regulate platelet reactive oxygen species (ROS) generation and subsequent platelet activation by mediating assembly of a functional NADPH oxidase. The authors did not directly measure platelet NADPH oxidase activity, however, which raises the possibility that VPS34 may influence platelet ROS generation and aggregation through a NADPH oxidase-independent pathway.

The role of NADPH oxidase in ROS-dependent platelet activation and aggregation is controversial.^2–4 Platelets are known to express NADPH oxidases,^1–8 which assemble from cytoplasmic subunits (p47phox, p67phox, p40phox, Rac1/2) and membrane subunits (Nox2-P22phox).⁹ Several studies have demonstrated that apocynin, a non-specific inhibitor of NADPH oxidase subunit assembly, prevents platelet ROS generation and aggregation,^{6, 10} and Delaney et al⁴ recently reported that deficiency of Nox2 within platelets leads to decreased superoxide generation and aggregation. However, other studies have questioned the importance of NADPH oxidase in platelet ROS production, activation and aggregation. For example, Dharmarajah and colleagues² observed that platelets isolated from mice lacking NADPH oxidase subunits such as Nox2 or p47Phox exhibited aggregation responses to thrombin and collagen that were similar to platelets from wild type mice. Additionally, they showed that apocynin inhibited platelet aggregation to a similar extent in platelets from wild type and NADPH oxidase-deficient mice.² Similar results were obtained by Walsh et al,³ who found that platelets from Nox2-deficient mice were not impaired in platelet ROS generation, activation or aggregation responses. The explanation for these discrepant findings in murine models is not yet understood, but it is noteworthy that platelets from human patients with chronic granulomatous disease (a genetic disorder characterized by deficiency of Nox2) also exhibit normal aggregation responses.^{11, 12} In light of these observations, it is possible that loss of VPS34 affects platelet ROS generation and aggregation responses through pathways that are independent from NADPH oxidase assembly.

As noted by Liu et al, both VPS34 and NADPH oxidases may represent attractive targets for antithrombotic therapeutics. Therefore, it will be important in future studies to define the specific downstream pathways of VPS34 signaling leading to platelet activation, perhaps by using genetic or pharmacological approaches to inhibit VPS34 in platelets from mice deficient in Nox2 and other NADPH oxidase subunits.