Serum angiotensin type-1 receptor autoantibodies and neurofilament light chain as markers of neuroaxonal damage in post-COVID patients

Abstract

Introduction: Dysregulation of autoimmune responses and the presence of autoantibodies (AA), particularly those related to the renin-angiotensin system (RAS), have been implicated in the acute phase of COVID-19, and persistent dysregulation of brain RAS by RAS-related autoantibodies may also contribute to neurological symptoms of post-COVID.

Methods: We analyzed levels of serum and CSF RAS AA in post-COVID patients with neurological symptoms, individuals who have fully recovered from COVID-19 (after-COVID controls), and uninfected individuals, and their possible correlations with the serum marker of neuroaxonal damage neurofilament light chain (NfL) and the degrees of cognitive deficit.

Results: Both in serum and CSF, levels of AA agonists of the pro-inflammatory angiotensin II type 1 receptors (AT1-AA) were significantly elevated in this cohort of neurological post-COVID patients compared to both uninfected and after-COVID controls and correlated with serum levels of NfL. Changes in serum and CSF levels of AA promoting the RAS anti-inflammatory axis (upregulation of AA agonists of AT2 and Mas receptors, downregulation of AA antagonists of ACE2) suggest upregulation of the RAS compensatory response in this cohort of neurological post-COVID patients. Post-COVID patients with more pronounced cognitive impairment exhibited significantly higher CSF levels of MasR-AA and a trend toward elevated AT2-AA. Persistent brain RAS dysregulation, particularly persistent increase in AT1-AA, and its correlation with neuroaxonal damage markers and cognitive impairment, may play a significant role in neurological symptoms associated with post-COVID. Serum levels of NfL and AT1-AA may be interesting biomarkers for the early identification of CNS involvement in patients with neurological symptoms and a history of COVID-19. However, post-COVID is a highly heterogeneous entity and may result from various underlying mechanisms. The present study includes a cohort, which may differ from other cohorts with different clinical profiles, which may show different results on NfLs and CSF RAS autoantibodies, particularly AT1-AA.

Conclusion: These findings highlight the potential of targeting AT1 receptors as a therapeutic strategy for mitigating cognitive deficits in post-COVID patients showing upregulated AT1-AA levels.

Keywords: AT1; COVID-19; autoantibody; autoimmunity; biomarkers; cognitive impairment; neurological long-COVID; post-acute sequelae of COVID-19 syndrome.

Figure 1

Schematic organization of the renin-angiotensin system (RAS), which is formed by two opposite axes: a first axis (red arrows) mainly constituted by Angiotensin II acting on AT1 receptors (AT1R), and a compensatory axis (green arrows) constituted by Angiotensin II acting on AT2 receptors, and Angiotensin 1-7 acting on Mas receptors. Angiotensin II is produced by the enzyme prorenin/renin acting on the precursor protein angiotensinogen to produce Angiotensin I, which is converted by the angiotensin-converting enzyme (ACE) into Angiotensin II. Renin and its precursor prorenin (PR) can also act on specific PR receptors. Angiotensin-converting enzyme 2 (ACE2) plays a key role in the system balance because ACE2 (together with peptidases such as Neprilysin, NEP) transforms components of the pro-inflammatory axis (Angiotensin I and, particularly, Angiotensin II) into components of the compensatory axis (Angiotensin 1-9 and, particularly Angiotensin 1-7). Autoantibodies (AA) play an agonistic effect on the GPCR (G protein-coupled receptors: AT1, AT2, Mas) and antagonistic effect on ACE2. Figure modified with permission from: Drugs Modulating Renin-Angiotensin System in COVID-19 Treatment. J.L. Labandeira-Garcia et al., 2022, Biomedicines, 10 (DOI:10.3390/biomedicines10020502).

Figure 2

Serum levels of different RAS autoantibodies in the three studied groups. (A) AT1-AA serum levels were significantly higher in the post-COVID patient group. Changes in autoantibody levels for the RAS anti-inflammatory axis [i.e. AT2 (B) and MasR (C)] were less clear. (D) Serum levels of ACE2-AA were significantly higher in the after-COVID control group. Data distribution is shown using a box plot with boxes representing the IQR, i.e., the limits between the first quartile (Q1, 25%) and the third quartile (Q3, 75%). The whiskers extend up to maximum and minimum values. *p < 0.05 relative to uninfected control; ^# p<0.05 relative to after-COVID controls. Kruskal–Wallis One Way Analysis of Variance on Ranks with Dunn’s Method post hoc test. ACE2-AA, ACE2 Autoantibodies; AT1-AA, Autoantibodies for AT1 receptors; AT2-AA, Autoantibodies for AT2 receptors; IQR, Interquartile range; MasR-AA, Autoantibodies for Mas receptors.

Figure 3

NfL serum levels in the three studied groups. (A) NfL serum levels were significantly higher in patients who had suffered COVID-19 but did not present post-COVID symptoms (after-COVID controls) and post-COVID patients than in uninfected controls and NfL levels in post-COVID patients were significantly higher than in after-COVID controls. Furthermore, levels of serum NfL significantly correlated with serum levels of AT1-AA in the entire study population (B), uninfected controls (C); after-COVID controls (D), and post-COVID patients (E). In (A), data distribution is shown using a box plot with boxes representing the IQR, i.e., the limits between the first quartile (Q1, 25%) and the third quartile (Q3, 75%). The whiskers extend up to maximum and minimum values. *p < 0.05 relative to uninfected control; ^# p<0.05 relative to after-COVID controls. Kruskal–Wallis One Way Analysis of Variance on Ranks with Dunn’s Method post hoc test. In (B–D), correlations between NfL serum levels and AT1-AA serum levels were assessed using Spearman’s rank correlation coefficient. AT1-AA, Autoantibodies for AT1 receptors; IQR, Interquartile range; NfL, Neurofilament light chain.

Figure 4

CSF levels of different RAS autoantibodies in the three studied groups. (A) CSF AT1-AA levels were significantly higher in the post-COVID patients. (B) CSF levels of AT2-AA were significantly higher in post-COVID patients than in uninfected patients but not significantly different compared to after-COVID controls. (C) In the post-COVID patient group, CSF MasR-AA levels exhibited a trend toward being higher than in the other groups, although this difference did not reach statistical significance. (D) CSF ACE2-AA levels were significantly lower in the post-COVID patient group than in the after-COVID control group. Data distribution is shown using a box plot with boxes representing the IQR, i.e., the limits between the first quartile (Q1, 25%) and the third quartile (Q3, 75%). The whiskers extend up to maximum and minimum values. *p < 0.05 relative to uninfected controls; ^# p<0.05 relative to after-COVID controls. Kruskal–Wallis One Way Analysis of Variance on Ranks with Dunn’s Method post hoc test. ACE2-AA, ACE2 Autoantibodies; AT1-AA, Autoantibodies for AT1 receptors; AT2-AA, Autoantibodies for AT2 receptors; IQR, Interquartile range; MasR-AA, Autoantibodies for Mas receptors.

Figure 5

CSF levels of different RAS autoantibodies in post-COVID patients with different degrees of cognitive impairment. No significant differences in CSF AT1-AA or ACE2-AA levels were found between the three post-COVID subgroups: unimpaired overall CI (z > -0.71), mild overall CI (-0.71 ≥ z >-1.40), and severe overall CI (z ≤ -1.40), (A, D). However, the group with more marked CI exhibited significantly higher CSF MasR-AA levels (C) and a trend toward higher CSF AT2-AA levels (B), though the latter did not reach statistical significance. Data distribution is shown using a box plot with boxes representing the IQR, i.e., the limits between the first quartile (Q1, 25%) and the third quartile (Q3, 75%). The whiskers extend up to maximum and minimum values. *p < 0.05 relative to ACE III (Average score) of Z> -0.71; ^# p<0.05 relative to ACE III (Average score) of -071<Z>-1.40. Kruskal–Wallis One Way Analysis of Variance on Ranks with Dunn’s Method post hoc test. ACE2-AA, ACE2 Autoantibodies; AT1-AA, Autoantibodies for AT1 receptors; AT2-AA, Autoantibodies for AT2 receptors; CI, Cognitive Impairment; IQR, Interquartile range; MasR-AA, Autoantibodies for Mas receptors.

Figure 6

Autoantibody correlations in post-COVID patients. Results of Spearman’s correlation matrixes show variables (AT1-AA, ACE2-AA, AT2-AA, MasR-AA and NfL) correlating at varying levels of significance (*p < 0.05; **p < 0.01; ***p < 0.001) in serum (A), CSF (B) and between serum and CSF (C). AT1-AA, Autoantibodies for AT1 receptors; ACE2-AA, ACE2 Autoantibodies; AT2-AA, Autoantibodies for AT2 receptors; MasR-AA, Autoantibodies for Mas receptors; NfL, Neurofilament light chain.