Background/Purpose: Patients with COVID-19 are at high risk for occlusion of vascular beds of all sizes. Considering endothelial cell activation has regularly been described as part of the COVID-19 thrombo-inflammatory storm, we aimed to characterize potential upstream mediators of this activation including neutrophil extracellular traps (NETs) and antiphospholipid antibodies (aPL).

Methods: Cultured human umbilical vein endothelial cells (HUVECs) were exposed to sera from 118 patients hospitalized with COVID-19. Plasma samples from 100 non-COVID sepsis patients were also characterized alongside 72 COVID samples. Upregulation of surface cell adhesion molecules E-selectin, ICAM-1, and VCAM-1 was determined by in-cell ELISA. Soluble E-selectin was measured in serum. HUVECs were also stimulated with IgG purified from three COVID patients positive for anticardiolipin IgG and separately five positive for anti-PS/PT IgG.

Results: As compared with sera from 38 healthy controls, COVID sera triggered an activated HUVEC phenotype as evidenced by markedly increased surface expression of the cell adhesion molecules E-selectin (mean 1.67-fold increase, p< 0.0001), VCAM-1 (mean 1.89-fold increase, p< 0.0001), and ICAM-1 (mean 1.66-fold increase, p< 0.0001). While non-COVID sepsis plasma elicited higher expression of surface ICAM-1 than did control plasma, the effect was even more robust with COVID-19 plasma (p< 0.05 comparing COVID-19 to sepsis). Beyond the in-cell ELISA platform, we also found significantly higher levels of soluble E-selectin in COVID serum as compared with healthy controls (p< 0.0001), where it correlated with clinical parameters that track with COVID-19 severity including C-reactive protein (r=0.31, p=0.005), D-dimer (r=0.31, p=0.008), calprotectin (r=0.29, p=0.003), and oxygenation efficiency (r=-0.31, p=0.002). One marker of NET remnants, myeloperoxidase-DNA complexes, modestly correlated with the ability of serum to increase expression of both VCAM-1 (r=0.26, p< 0.01) and ICAM-1 (r=0.28, p< 0.01) on HUVECs, while cell-free DNA and citrullinated histone H3 (additional markers of NETs) correlated with VCAM-1 only (r=0.24, p< 0.05 and r=0.22, p< 0.05, respectively). Interestingly, we detected robust correlations between various aPL in serum, especially anticardiolipin IgG/M and anti-phosphatidlyserine/prothrombin (anti-PS/PT) IgG/M, and the three markers of HUVEC activation (E-selectin, VCAM-1, and ICAM-1) (Table 1). As compared with mock depletion, IgG depletion strongly abrogated the ability of pooled COVID sera (anticardiolipin-positive and anti-PS/PT-positive) to upregulate HUVEC E-selectin (p< 0.01 and p< 0.05, respectively), VCAM-1 (p< 0.05 and p< 0.01) and ICAM-1 (p< 0.01 for both). Furthermore, IgG (100 μg/ml) purified from the anticardiolipin-positive and anti-PS/PT-positive COVID serum increased expression of ICAM-1 (p< 0.01 for both) when spiked into control serum.

Conclusion: These data are the first to suggest that patient antibodies are a driver of endothelial cell activation in COVID-19 and add important context regarding thrombo-inflammatory effects of aPL-like autoantibodies in severe COVID-19.

« Back to ACR Convergence 2021

ACR Meeting Abstracts - https://acrabstracts.org/abstract/endothelial-cell-activating-antibodies-in-covid-19/