Endothelial Colony-Forming Cells (ECFC) As an Autologous Model for Studying Endothelial Pathophysiology in Sickle Cell Anemia and Myeloproliferative Neoplasms

Background: Endothelial colony-forming cells (ECFC) are an important source of autologous endothelial cells to study its implication in the pathophysiology of diseases with risk of vaso-occlusive events. Currently, our research group began to elucidate the ECFC mechanisms that contribute to the complex clinical vascular manifestations in two diseases, sickle cell anemia (SCA) and myeloproliferative neoplasms (MPN).

Aims: In this study, we analyzed functional in vitro assays of endothelial cells. The adhesion to red blood cell (RBC), migration and angiogenesis process of ECFC isolated from patients with SCA and MPN, as well as from healthy individuals (CTR) were evaluated in seeking an expanded understanding of the biology of the endothelial cell and its role in vascular events.

Methods: ECFC were obtained through the isolation and culture of human peripheral blood mononuclear cells. ECFC were isolated from 8 patients with SCA under regular transfusion, 6 patients with MPN and 10 CTR. Human umbilical vein endothelial cells (HUVEC) were used as additional controls. Flow cytometry of ECFC indicated that all cells were highly positive for endothelial cell markers CD31, CD144, CD146 and KDR with no indication of CD45 (leukocyte antigens), CD133 (endothelial progenitor cell marker) and CD34 (hematopoietic progenitor cell marker). RBCs from healthy individuals were obtained after centrifugation of whole blood. Cellular adhesion was evaluated after incubation of ECFCs with RBCs in the presence or absence of inflammatory stimulus (TNF-α). Endothelial adhesion molecules were analyzed by flow cytometry (ICAM-1, VCAM-1, E and P-selectin). ECFC migration was assessed using a scratch-wound healing assay and wound regression was analyzed by time-lapse videos. Angiogenesis capacity was evaluated through three-dimensional ECFC cultures in Matrigel. Network parameters (segments, junctions and meshes) were characterized during 24h after seeding. All experiments were performed in triplicate.

Results: In total, 48 ECFC colonies were established, 10 from SCA, 25 MPN and 13 CTR. We observed a higher percentage of adhered RBCs to ECFCs isolated from patients with SCA (14.0%) and MPN (23.4%) without TNF-ɑ stimulus, when compared to ECFC from CTR, (8.4%, p< 0.04 and p< 0.001, respectively), and HUVEC (9.4%, p< 0.05 and p< 0.02, respectively). No differences were detected in the expression of adhesion molecules in ECFC.

Mean wound regression rates at 14h were 79.9% for SCA, 84.4% for MPN, and 88.8% for CTR. The high variability among colonies in each group could explain why this difference was not statistically significant. Finally, HUVECs had a shorter time for wound closure, with complete wound regression at 10h.

The angiogenesis analysis at 15h ECFC from SCA and MPN had, respectively, 20% and 50% less network parameters than CTR. Then beyond 15h post-seeding, the network parameters regressed until reaching a plateau. At 24h the segments began to disappear progressively, leading to a marked reduction in 40h. Unlike ECFC, HUVEC presented a high network formation.

Conclusions: Our findings reveal distinct functional properties and behavior between the ECFCs from two diseases with vascular manifestations, SCA and MPN. ECFC do provide access to patient vascular endothelium and enable us to validate the use of these cells as investigative models. In contrast, HUVECs showed a unique behavior, which differed from both diseased and controls ECFCs. This highlights the differences between autologous in vitro and established cell lines as experimental models for vascular diseases. This raises the question of which is the most representative model of human endothelial pathophysiology in vascular diseases.