Autoantibodies against ACE2 and angiotensin type-1 receptors increase severity of COVID-19

Abstract

The renin-angiotensin system (RAS) plays a major role in COVID-19. Severity of several inflammation-related diseases has been associated with autoantibodies against RAS, particularly agonistic autoantibodies for angiotensin type-1 receptors (AA-AT1) and autoantibodies against ACE2 (AA-ACE2). Disease severity of COVID-19 patients was defined as mild, moderate or severe following the WHO Clinical Progression Scale and determined at medical discharge. Serum AA-AT1 and AA-ACE2 were measured in COVID-19 patients (n = 119) and non-infected controls (n = 23) using specific solid-phase, sandwich enzyme-linked immunosorbent assays. Serum LIGHT (TNFSF14; tumor necrosis factor ligand superfamily member 14) levels were measured with the corresponding assay kit. At diagnosis, AA-AT1 and AA-ACE2 levels were significantly higher in the COVID-19 group relative to controls, and we observed significant association between disease outcome and serum AA-AT1 and AA-ACE2 levels. Mild disease patients had significantly lower levels of AA-AT1 (p < 0.01) and AA-ACE2 (p < 0.001) than moderate and severe patients. No significant differences were detected between males and females. The increase in autoantibodies was not related to comorbidities potentially affecting COVID-19 severity. There was significant positive correlation between serum levels of AA-AT1 and LIGHT (TNFSF14; r_Pearson = 0.70, p < 0.001). Both AA-AT1 (by agonistic stimulation of AT1 receptors) and AA-ACE2 (by reducing conversion of Angiotensin II into Angiotensin 1-7) may lead to increase in AT1 receptor activity, enhance proinflammatory responses and severity of COVID-19 outcome. Patients with high levels of autoantibodies require more cautious control after diagnosis. Additionally, the results encourage further studies on the possible protective treatment with AT1 receptor blockers in COVID-19.

Keywords: Autoantibody; Autoimmunity; LIGHT; Outcome prediction; Renin-angiotensin system; SARS-CoV-2.

Graphical abstract

Fig. 1

Levels of AA-AT1 and AA-ACE2 in controls and COVID-19 patients. COVID-19 patients had significantly higher serum levels of AA-AT1 (A; two-way ART-ANOVA, p = 0.002) and AA-ACE2 (B; two-way ART-ANOVA, p = 0.013) than the control group. However, no significant differences were detected between males and females for serum levels of AA-AT1 (two-way ART-ANOVA, p = 0.845) and AA-ACE2 (two-way ART-ANOVA, p = 0.342). Interaction terms were not significant for both levels of antibodies. Data distribution is shown using a box plot with boxes representing the IQR and the median (black line) and whiskers representing ±1.5 IQR. IQR: Interquartile range.

Fig. 2

Levels of AA-AT1 and AA-ACE2 in patients with different levels of COVID-19 disease severity. Mild disease patients had significantly lower levels of AA-AT1 (A) and AA-ACE2 (B) than moderate and severe patients (Kruskal–Wallis one-way analysis of variance on ranks followed by Wilcoxon test). In the present study, no significant differences were detected between females and males for serum levels of AA-AT1 (C; two-way ART-ANOVA, p = 0.95) and AA-ACE2 (D; two-way ART-ANOVA, p = 0.14) in any of the severity levels of the disease. Data distribution is shown using a box plot with boxes representing the IQR and de median (black line) and whiskers representing ±1.5 IQR. IQR: Interquartile range.

Fig. 3

Levels of AA-ACE2 in COVID-19 patients with and without diabetes and correlation between AA-AT1 levels and LIGHT levels. AA-ACE2 levels (A) were significantly higher in patients with diabetes than in non-diabetic patients. Even after controlling the possible confounding effect of diabetes, the relationship of antibodies and disease severity was still significant. Data distribution is shown using a box plot with boxes representing the IQR and the median (black line) and whiskers representing ±1.5 IQR. *p < 0.05 relative to non-diabetic group (Wilcoxon-Mann-Whitney test). IQR: Interquartile range. AA-AT1 levels positively correlated with LIGHT levels (B). Scatterplot showing positive and linear association (r_Pearson = 0.83, CI_95% = [0.74, 0.89]; p < 0.001) between AA-AT1 levels and LIGHT levels. The distribution of the variables is shown in the histograms. CI: Confidence Interval.

Fig. 4

Proposed model for AA-AT1 and AA-ACE2 effects. SARS-CoV-2 infection induces increase in pro-inflammatory cytokines, particularly LIGHT, promoting AA-AT1, which act as AT1 receptor agonists and enhance the pro-inflammatory RAS axis. SARS-CoV-2 binds cell surface ACE2 leading to a decrease in this transmembrane ACE2 and an increase in levels of soluble/circulating ACE2. A decrease in transmembrane ACE2 further enhances the pro-inflammatory RAS axis and reduces anti-inflammatory axis activity. The increase in levels of circulating ACE2-SARS-CoV-2 complexes may increase levels of AA-ACE2, which further reduce transmembrane ACE2 activity and the anti-inflammatory RAS function. Green lines, beneficial effects; red lines, detrimental effects.