PNH is an acquired clonal disorder of hematopoietic stem cells due to a somatic mutation of the PIG-A gene, resulting in a membrane deficiency of glycosylphosphatidyl inositol-anchored proteins (GPI-AP). The mechanism of thrombosis, the most serious complication of PNH, has not been sufficiently clarified. GPI-AP functions include enzymatic activities and maintenance of lipid rafts (LR) integrity. Some GPI-AP also serve as a scaffold for various proteins. Therefore deficiency of GPI-AP has consequences for the expression of some non-GPI-AP. For example, GPI-anchored NB1 (CD177) glycoprotein co-localizes with membrane-bound proteinase 3 (mPR3) on granulocytes. We demonstrate that anti-CD177 antibody co-precipitated PR3 and the PNH phenotype results in a concomitant loss of mPR3 expression despite the fact that PR3 does not have a GPI tail. Among other functions, PR3 can modulate thrombus formation by cleavage of the thrombin receptor (TR) downstream from the thrombin cleavage site, decreasing thrombin-mediated platelet activation. We show that pre-incubation of platelets with purified PR3 resulted in a significant inhibition of platelet activation in response to thrombin, as measured by flow cytometric analysis of P-selectin and PAC-1 expression, while PR3 itself did not affect platelet function. However, PR3 failed to inhibit platelet activation induced by ADP and/or TRAP42-55 agonist, suggesting that PR3 renders TR inactive to thrombin while thrombin-independent activation is not affected. Membrane deficiency of PR3 on granulocytes may contribute to unopposed platelet activation promoting a procoagulable state inherent to PNH. Consequently, we investigated PR3 expression in PNH; when we applied surface/intracellular cytometry and Western blotting for PR3 in patients (N=20), mPR3 was present on the surface of wild-type but absent on GPI-deficient granulocytes. However, intracellular levels of PR3 were equal in both cell types. Association of CD177 and mPR3 was further supported by analysis of LR purified from PNH granulocytes. Immunoblots preformed following density gradient separation, showed a complete loss of LR-associated PR3 and CD177 in PNH cells while PR3 was present in cytoplasmatic fractions. Similar observation was made when another related neutrophilic protease, cathepsin G (CTSG) was studied and found in cytoplasmic fractions in PNH cells, but was detected only in LR derived from normal cells. When we tested the levels of plasma PR3 in patients with PNH (N=40) and controls (N=69), we demonstrated that secreted PR3 is derived from the membrane bound fractions as circulating levels of PR3 were significantly decreased in PNH patients when compared to healthy and hematologic controls with unrelated conditions (p=.0023 and p<.0001). Thus the decrease in the overall levels of available PR3 can be related not only to lower membrane but also circulating PR3 levels in PNH. However, when we tested either probands with NB1/PR3-null phenotype or patients with lower NB1/PR3 membrane expression, normal or elevated level of sPR3 was found, likely compensating for membrane PR deficiency. Our study demonstrates a novel mechanism that may be contributing to the increased thrombotic risk in PNH. Among PNH patients who experienced thrombotic complications 7/9 showed decreased sPR3 and significantly lower level of mPR3 than patients without history of thrombotic events.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author