Role of Protein Disulfide Isomerase (PDI) In the Expression of Tissue Factor (TF) Procoagulant Activity on Apoptotic Myeloblasts

Abstract 152

Disseminated intravascular coagulation (DIC) is a common and potentially life-threatening complication of acute myelogenous leukemia (AML). Although various components of the procoagulant and fibrinolytic systems have been implicated in its pathogenesis, aberrant expression of TF appears to be critically involved. Despite the need for more specific hemostatic agents, the molecular pathways underlying the expression of TF procoagulant activity (PCA) in AML remain obscure. In this study, we sought to identify determinants for systemic coagulation activation in a prospective cohort of 50 patients with newly diagnosed AML and conducted a series of in-vitro experiments to further characterize TF PCA regulation on myeloblasts in light of recently provided evidence. AML patients had significantly elevated plasma D-dimer and TF-specific PCA of isolated peripheral blood mononuclear cells (PB-MNCs) compared to healthy controls (n=10) and patients with ALL (n=5) or MDS (n=7), demonstrating specific upregulation of the TF-dependent coagulation pathway. In the AML cohort, TF PCA of PB-MNCs was 30-fold higher and TF+ plasma microparticles were more frequently detectable in patients with hypofibrinogenemia and a thrombohemorrhagic syndrome (n=8) as compared to patients with normal fibrinogen levels (n=42), providing further evidence that TF over-expression by myeloblasts is indeed a determinant for DIC decompensation. Physical cell disruption resulted in a 3-fold increase in TF PCA of PB-MNCs, indicating that a significant proportion of TF was non-coagulant (i.e. cryptic) on unperturbed cells. In patients with AML, TF PCA of PB-MNCs, absolute numbers of circulating blasts and lactate dehydrogenase (LDH) serum levels, a surrogate marker for leukemic cell turnover, were significantly correlated with plasma D-dimer, suggesting that membrane alterations associated with apoptosis/necrosis played a crucial role in myeloblast TF PCA expression. Extracellular PDI, which is abundantly expressed in the endoplasmic reticulum, has recently been implicated in TF regulation via its oxidoreductase and chaperone activity. In monoblastic THP1 cells harvested at varying time points during maintenance culture (n=5), we found a significant correlation between membrane PDI expression, as assessed by flow cytometry, and cell-associated TF PCA (r=0.93). Analysis of annexin V-FITC binding revealed that PDI was predominantly expressed on apoptotic monoblasts. Treatment of THP1 cells for 24 h with 1 μ M daunorubicin (DNR), a cytotoxic drug commonly used in AML therapy, induced >80% apoptosis and increased cellular TF PCA 8 ± 3-fold (n=8) independent of de-novo protein synthesis. Consistently, apoptotic cells stained extensively positive for PDI by flow cytometry. DNR-induced TF PCA was inhibited by 70 ± 6% (n=8) when THP1 cells were coincubated with RL90, an inhibitory PDI monoclonal antibody previously shown to diminish fibrin and platelet deposition in murine models of micro- and macrovascular thrombosis. Importantly, RL90 had no effect when added to THP1 cells at the end of the 24-h culture period, indicating specific effects on the expression of cellular TF PCA rather than the coagulation reaction itself. Similarly, coincubation of DNR-treated THP1 cells with the PDI inhibitor bacitracin dose-dependently (1-5 mM) inhibited TF activation by up to 93 ± 10% (n=3), while only a 30 ± 16% inhibition was observed when bacitracin was added at the end of the culture period. Virtually identical results were obtained with the AML cell lines, HL60 and U937. Furthermore, ex-vivo treatment of isolated myeloblasts from a patient with acute promyelocytic leukemia and decompensated DIC with 1 μ M DNR for 24 h induced >90% apoptosis and increased cellular TF PCA 5-fold. Extensive PDI staining was observed on apoptotic myeloblasts, and coincubation with bacitracin (5 mM) completely inhibited DNR-induced upregulation of TF PCA while not affecting baseline TF PCA expression. Finally, treatment with DNR resulted in the leakage of both LDH and PDI into the culture supernatants of AML cell lines, and soluble PDI antigen was detectable in the plasma samples of AML patients with the highest LDH serum levels. In summary, our findings suggest that, in addition to negatively charged phospholipids, surface-expressed and/or released PDI may be required for the efficient expression of TF-specific PCA on apoptotic myeloblasts.