Sustained VWF-ADAMTS-13 axis imbalance and endotheliopathy in long COVID syndrome is related to immune dysfunction

Abstract

Background: Prolonged recovery is common after acute SARS-CoV-2 infection; however, the pathophysiological mechanisms underpinning Long COVID syndrome remain unknown. VWF/ADAMTS-13 imbalance, dysregulated angiogenesis, and immunothrombosis are hallmarks of acute COVID-19. We hypothesized that VWF/ADAMTS-13 imbalance persists in convalescence together with endothelial cell (EC) activation and angiogenic disturbance. Additionally, we postulate that ongoing immune cell dysfunction may be linked to sustained EC and coagulation activation.

Patients and methods: Fifty patients were reviewed at a minimum of 6 weeks following acute COVID-19. ADAMTS-13, Weibel Palade Body (WPB) proteins, and angiogenesis-related proteins were assessed and clinical evaluation and immunophenotyping performed. Comparisons were made with healthy controls (n = 20) and acute COVID-19 patients (n = 36).

Results: ADAMTS-13 levels were reduced (p = 0.009) and the VWF-ADAMTS-13 ratio was increased in convalescence (p = 0.0004). Levels of platelet factor 4 (PF4), a putative protector of VWF, were also elevated (p = 0.0001). A non-significant increase in WPB proteins Angiopoietin-2 (Ang-2) and Osteoprotegerin (OPG) was observed in convalescent patients and WPB markers correlated with EC parameters. Enhanced expression of 21 angiogenesis-related proteins was observed in convalescent COVID-19. Finally, immunophenotyping revealed significantly elevated intermediate monocytes and activated CD4+ and CD8+ T cells in convalescence, which correlated with thrombin generation and endotheliopathy markers, respectively.

Conclusion: Our data provide insights into sustained EC activation, dysregulated angiogenesis, and VWF/ADAMTS-13 axis imbalance in convalescent COVID-19. In keeping with the pivotal role of immunothrombosis in acute COVID-19, our findings support the hypothesis that abnormal T cell and monocyte populations may be important in the context of persistent EC activation and hemostatic dysfunction during convalescence.

Keywords: Weibel Palade body exocytosis; convalescent COVID-19; endothelial cell activation; immune dysfunction; long COVID.

FIGURE 1

Comparisons are shown among acute COVID‐19 patients (n = 36), convalescent COVID‐19 patients (n = 50), and healthy controls (n = 20) including: (A) plasma ADAMTS‐13:Ag levels, (B) plasma VWF/ADAMTS‐13 ratios, (C) plasma Platelet Factor 4 (PF4) levels, and (D) plasma Interleukin‐6 (IL‐6) levels. Dotted red lines denote the lower limit of the local reference range for plasma ADAMTS‐13:Ag and IL‐6 levels and the upper limit of the local reference range for the VWF/ADAMTS‐13 ratio and PF4 levels. Data are presented as median and the interquartile range. Comparisons between groups were assessed using the Kruskal‐Wallis test (ns = not significant, **p <0 .01,***p < 0.001, ****p < 0.0001).

FIGURE 2

Comparisons are shown between acute COVID‐19 patients (n = 36), convalescent COVID‐19 patients (n = 50), and healthy controls (n = 20) including: (A) plasma Angiopoietin‐2 (Ang‐2) levels, and (B) plasma Osteoprotegerin (OPG) levels. Dotted red lines denote the upper limit of the local reference range for plasma Ang‐2 and OPG. Data are presented as median and the interquartile range. Comparisons between groups were assessed using the Kruskal‐Wallis test. Correlations are shown between plasma levels of OPG and endothelial cell activation parameters including: (C) von Willebrand factor propeptide (VWFpp), (D) Factor VIII:C, and (E) Angiopoietin‐2. Correlations were evaluated using the Spearman rank correlation test. (F) Heatmap visualization indicating angiogenesis‐related protein expression detected in each subject (columns) for each protein (rows), comparing a subset of convalescent COVID‐19 patients (n = 9, 6/9 males) and healthy controls (n = 3, 2/3 males). This subset of convalescent patients and controls were selected to match the age and sex profile of the overall convalescent cohort. Protein levels were measured via membrane‐based antibody array and data are represented by mean pixel intensity with red indicating higher and green indicating lower levels of the protein of interest. Comparisons between groups were assessed using the Mann–Whitney U test.(ns = not significant, ***p <0 .001, ****p <0 .0001).

FIGURE 3

(A) Intermediate monocytes in peripheral blood were assessed using flow cytometry in convalescent COVID‐19 patients (n = 37), acute COVID‐19 patients (n = 32), and healthy controls (n = 20). Correlations between intermediate monocytes and (B) endogenous thrombin potential (ETP) and (C) peak thrombin generation, respectively. T lymphocyte subsets in peripheral blood were assessed using flow cytometry in convalescent COVID‐19 patients (n = 37), acute COVID‐19 patients (n = 32), and healthy controls (n = 20) including percentage of (D) naïve CD4, (E) naïve CD8, (F) activated CD4, and (G) activated CD8 positive T cells. Correlations are shown between activated CD8+ T cells and plasma levels of (H) soluble thrombomodulin (sTM), (I) Osteoprotegerin (OPG), and (J) VWF:Ag. Data are presented as median and the interquartile range. Comparisons between groups were assessed by the Mann–Whitney U test. Correlations were evaluated using the Spearman rank correlation test (ns = not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).