Introduction

Hemophilia A (HA) is associated with coagulation FVIII insufficiency or inactivity and leads to excessive bleeding. Elevated FVIII on the contrary is associated with risk of thrombosis and stroke. Active FVIII uses its C2 domain to bind to blood cells' membranes, consequently carrying out its coagulative function. Leukocyte abundance is reported to be associated with thrombotic events; however, this phenomenon lacks a clear mechanistic explanation.

We developed a reliable Flow-Cytometry (FC) method for FVIII detection that can be utilized for assessing the presence of FVIII in the cytoplasm and on the surface of leukocytes. Here, we tested through FC the hypothesis that FVIII bound on leukocytes may be relevant in different coagulation states.

Methods

We analyzed a cohort of 44 pediatric subjects, median age 5 years, including patients with HA (n=12), other coagulopathies (von Willebrand Disease (vWD)/hemophilia B/unexplained menorrhagia/bleedings (n=11), venous thrombosis (n=8) and healthy controls (n=13), to investigate if the amount of FVIII bound on leukocytes could be associated with their different coagulation/clinical state.

Blood samples were selected for high quality and stained with a fixable viability staining, CD3, CD4, CD8, CD19, CD14, CD33, CD16 and anti-FVIII antibodies (clone GMA8024, Green Mountain Abs) utilizing Zenon technology as previously described (Elnaggar, Mol Ther Methods Clin Dev. 2019) but only targeting the cell surface, without permeabilizing the cells prior to FVIII staining.

Results

Total leukocytes' surface FVIII showed a clear declining trend across thrombosis, normal, and hypo-coagulation states. However, no significant correlation was observed between circulating levels of FVIII in plasma and the levels of FVIII bound to leukocytes, indicating that the differences in FVIII surface binding are not directly proportional to the availability of FVIII in the circulation and suggesting a specific binding mechanism governing the interaction between FVIII and leukocytes.

Interestingly, patients (n=4) with a more recent-onset thrombosis at time of sampling, receiving heparin-based therapy for only 1-3 days had a significant higher FVIII bound on total leukocytes with respect to patients (n=4) with 6 days-2 months of therapy (p=0.0229).

When analyzing the distinct blood subpopulations in different coagulation status, we observed that in general, leukocytes' surface FVIII was more concentrated on myeloid compared to lymphoid cells, and the most prominent difference was in classical monocytes of HA patients versus thromboses patients (p=0.0368).

Interestingly, monocytes are pathologically relevant for thrombus formation. They expose their cellular surfaces for coagulation factor complexes assembly, release tissue factor and (similar to neutrophils) produce extracellular traps that facilitate blood cells and coagulation factors (such as FVIII and vWF) adherence at the site of thrombus formation.

After observing the elevated levels of cellular surface FVIII on monocytes of thrombosis patients, we sought to test whether monocytes' surface FVIII levels were allowing to predict the patient's clinical coagulation state. Binomial regression analyses showed that monocyte's surface FVIII is powerful in predicting thrombosis (AUC=0.79) with respect to the other clinical states and not significantly predicting any other coagulation states.

Conclusions

We showed that peripheral blood monocytes have a specific high abundance of surface FVIII in thrombosis patients. Importantly, monocytes are a primary extrahepatic source of FVIII. This may suggest that monocytes produce and secrete FVIII and carry part of the secreted protein on their surface directing it into the thrombus formation site. Alternatively, they may actively bind active FVIII from the plasma pool and concentrate it at the thrombosis site.

Our study highlights the reliability of our FC platform in assessing FVIII abundance on leukocytes' membranes across coagulation states. Monocytes, particularly in cases of thrombosis, exhibit active binding of FVIII on their surface, suggesting a specific role in the pathophysiology of thrombosis that requires further investigation.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
Sign in via your Institution