Red Blood Cell-Derived Particles Induce Endothelial Damage Via the Alternative Complement Pathway

Introduction

Acute hyper-hemolysis is a severe and potentially life-threatening complication that can occur during delayed hemolytic transfusion reactions (DHTR) or treatment procedures such as Extracorporeal Membrane Oxygenation (ECMO). This pathological condition leads to increased plasma hemoglobin (Hb) levels and subsequent heme release from Hb oxidation, resulting in multi-organ failure. Our previous study, using a fluidic model and a humanized sickle cell disease mouse model reproducing hyper-hemolysis onset (Nguyen et al., Blood, 2024), suggested that hemolysate suspension, in the absence of heme, could induce endothelial and organ damage. In this study, we aim to identify the mechanisms involved in this process.

Materials and Methods

Washed red blood cells (RBCs) were reconstituted in compatible serum at 2.5% hematocrit and then sonicated to prepare whole hemolysate, which was characterized by spectrophotometry and flow cytometry.

Human Umbilical Vein Endothelial Cells (HUVECs) were exposed to serum alone or the same serum containing either whole hemolysate or various hemolysate components under flow conditions (1 dyn/cm²). In experiments requiring complement analysis, cell assays were performed under static conditions to avoid non-specific complement activation. Anti-C3 (Compstatin Cp40), and anti-C5 antibody (Eculizumab) were added in specific experiments to elucidate underlying mechanisms. Endothelial activation was assessed by measuring membrane expression of ICAM-1. Endothelial damage was evaluated by actin network staining, platelet thrombus formation, and RBC adhesion. Complement fragments and proteins was measured using Multiplex and EIA kits.

Results and Discussion

Hemolysate suspensions contained 522±29 µM free Hb, negligible MetHb (<0.1%), and no heme. Particles larger than 1 µm were significantly more abundant in hemolysate compared to serum (1.7-fold, p=0.002) whereas microparticles were similar in hemolysate and serum. These particles were predominantly CD235a positive (68%), with subpopulations co-expressing CD45 (3.6%), CD41 (7%), and CD31 (18%), and 65% were C3 positive. Hb binding to RBC-derived particles correlated with particle size. Centrifugation at 14,000 g effectively removed these particles. Both whole hemolysate and the 14,000 g pellet induced HUVEC activation and damage, but these effects were significantly reduced when using the supernatant of 14,000 g centrifugation or purified AA Hb at equivalent concentrations, indicating that RBC-derived particles play a critical role in endothelial activation during early hemolysis.

We quantified complement components in the supernatant of HUVECs treated with hemolysate to elucidate the role of complement activation in hemolysate-induced endothelial damage. The hemolysate supernatant showed higher levels of final common pathway proteins and fragments compared to serum: C5a and sC5b9 (1.6- and 1.3-fold, p=0.06 and 0.01, respectively). Elements related to the alternative pathway, including C3, C3a, Ba, and Bb were also increased (1.2, 1.8, 1.7, and 1.6-fold, respectively, p=0.06). However, no difference was observed in the C1q, C2, C5, factor I, D, P, H, and C4a levels. Heat-induced decomplementation of serum led to a significant reduction in hemolysate-induced endothelial activation and damage.

Treatment with either anti-C3 or anti-C5 Ab reduced endothelial activation associated with a diminution in the level of C5a (ANOVA vs control conditions, p=0.02 in two treatments) and sC5b9 (p=0.04 and 0.02 respectively) in a dose-dependent manner. As expected, C4a levels were unaffected by either treatment. In addition, anti-C3 Ab treatment significantly reduced the levels of alternative pathway fragments: C3a (p=0.001), Ba (p=0.001), and Bb (p=0.0008), whereas anti-C5 Ab had no impact on these fragments.

Conclusions and perspectives

These results suggest that, in the absence of heme, hemolysate containing RBC particles can trigger complement activation via the alternative pathway leading to endothelial activation and damage. This finding provides critical insight into the pathophysiology of acute hyper-hemolysis and highlights the importance of early treatment by complement inhibition in the onset phase. Ongoing research aims to reinforce these results and demonstrate complement deposition in cell-based or in vivo models.