key: cord-0284344-e0hv1x4b
authors: Harrington, W. E.; Jiang, Y.; Duffy, F.; Hadlock, J.; Raappana, A.; Styrchak, S.; Beck, I.; Chour, W.; Houck, J.; Duvvuri, V.; Yeung, W.; Haglund, M.; Wallner, J.; Wallick, J.; Hardy, S.; Oldroyd, A.; Ko, D.; Gervassi, A.; Murray, K.; Kaplan, H.; Aitchison, J.; Heath, J.; Sather, D. N.; Goldman, J.; Frenkel, L.
title: Angiotensin II receptor I auto-antibodies following SARS-CoV-2 infection
date: 2021-07-05
journal: nan
DOI: 10.1101/2021.06.30.21259796
sha: 44e266217995046e5f32b9212e53189ddf284f7c
doc_id: 284344
cord_uid: e0hv1x4b

BACKGROUND: Coronavirus disease 2019 (COVID-19) is associated with endothelial activation and coagulopathy, which may be related to pre-existing or infection-induced pro-thrombotic autoantibodies such as those targeting angiotensin II type I receptor (AT1R-Ab). METHODS: We compared prevalence and levels of AT1R-Ab in COVID-19 cases with mild or severe disease to age and sex matched negative controls. RESULTS: There were no significant differences between cases and controls. However, there were trends toward a higher proportion with AT1R-Ab positivity among severe cases versus controls (32% vs. 11%, p=0.1) and higher levels in those with mild COVID-19 compared to controls (median 9.5U/mL vs. 5.9U/mL, p=0.06). CONCLUSIONS: These findings suggest that AT1R-Ab are not consistently associated with COVID-19 but do not exclude a contribution to endothelial pathology in a subset of people.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. ; https://doi.org/10.1101/2021.06.30.21259796 doi: medRxiv preprint 5 antigen (HLA) typing by direct sequencing had previously been conducted for the INCOV 6 1 cohort. Cases were matched 1:1 to controls by age and sex. Healthy controls were derived from 6 2 the Children's SARS-CoV-2 Prospective Cohort. At the time of blood collection (March to July, 6 3 2020), controls reported no prior history of SARS-CoV-2-like illness, were nasal specimen PCR 6 4 negative for SARS-CoV-2, and were SARS-CoV-2 antibody negative on the SCoV-2 Detect™ 6 5

IgG and IgM ELISAs (InBios, Seattle, WA). All participants provided written informed consent. To assess the biologic plausibility of cross-reactive antibody binding sites, searches to find 6 8 potential regions of structural homology between AT1R and SARS-CoV-2 Spike protein were 6 9

performed. The UCSF Chimera MatchMaker tool was used, which aligns structures based on 7 0 pairwise sequence alignment followed by 3D structural alignment using residue Cα positions. The concentration of AT1R-Ab in plasma was measured with a quantitative ELISA (Celltrend, 7 5

Luckenwalde, Germany) using the entire AT1R protein, with assays performed according to the 7 6 manufacturer's instructions. In brief, standards and diluted samples (1:100) were added to 7 7 AT1R-precoated microtiter plates and incubated for 2 h at 4°C. After washing, AT1R-Ab was 7 8 detected with HRP-labelled anti-human IgG antibody followed by enzymatic substrate reaction.

Samples were tested in duplicate. Optical densities (OD) were converted into concentrations by 8 0 comparison to a standard curve and mean concentration was used to define a sample as CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 5, 2021. In order to investigate potential cross-reactivity between AT1R-Ab and SARS-CoV-2 Spike, we 8 5 tested plasma from AT1R-Ab positive control subjects and randomly selected AT1R-Ab 8 6 negative controls for reactivity against SARS-CoV-2 Spike trimer (including both S1 and S2 8 7 domains), as previously published (12). Because we hypothesized that potential cross-reactive 8 8 antibodies might bind SARS-CoV-2 Spike with lower avidity, we started with an initial plasma 8 9 dilution of 1:10 followed by serial dilutions by a factor of two. To assess for potential cross- control subjects against the receptor binding domain (RBD) in S1 and the nucleocapsid protein antibody responses were defined at a titer greater than or equal to 1:50 for each antigen. AT1R-Ab prevalence and levels in all COVID-19 cases were first compared to those in the 9 7 entire set of controls. In secondary analysis, the subsets with mild and severe COVID-19 were 9 8 separately compared to their age and sex matched controls. The distribution of AT1R-Ab status 9 9

(positive versus intermediate/negative) was compared between groups using the Chi-square 1 0 0 test, and AT1R-Ab level was compared between groups using quantile regression on the square test. Significance was defined as p≤0.05, whereas trend was defined as p≤0.1. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. There was no significant difference in the proportion of AT1R-Ab positive participants between vs. 21%, p=1) (Figure 1A, 1B) . However, there was a trend toward higher proportion with 1 1 4 AT1R-Ab positivity among severe cases versus controls (32% vs. 11%, p=0.1) (Figure 1C ). 1 1 5 Similarly, there was not a statistically significant difference in the median AT1R-Ab level 1 1 6 between any group, but a trend toward higher AT1R-Ab levels was observed in all cases versus cases and controls (median 9.5 vs. 5.9 U/mL, p=0.06), but not severe COVID-19 cases and versus 11%, p=0.1). Within the severe COVID-19 group, we tallied six participants with thrombotic events. These . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. ; https://doi.org/10.1101/2021.06.30.21259796 doi: medRxiv preprint 1 0 thrombosis (33% vs 31%, p=0.9). While the median level of AT1R-Ab was higher in those with 1 3 5 thrombosis, this was not statistically significant (12.5 vs. 5.7 U/mL, p=0.3). proportion of AT1R-Ab positive versus negative participants carried a DRB1*04 allele; however, 1 3 8 this was not statistically significant (50% vs. 18%, p=0.2) (Supplementary Table 1 ).

Structural homology between AT1R and SARS-CoV-2 Spike and assessment of antibody cross- Potential cross-reactivity of antibodies between SARS-CoV-2 Spike and AT1R was predicted second extracellular domain targeted by AT1R-AA (Figure 2) . To test the hypothesis that AT1R- Ab cross-reacts against SARS-CoV-2 Spike, in particular the S2 domain, the nine AT1R-Ab CoV-2 Spike trimer by ELISA. AT1R-Ab status was not associated with reactivity to SARS-CoV- 2 Spike trimer (17% AT1R-Ab negative vs. 22% positive control participants, p=0.8). In addition, SARS-CoV-2 Spike (S1 and S2) and AT1R were aligned to assess potential homology. Overlap . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 1 1 CoV-2 positive cases. Endpoint titers were defined as the reciprocal of plasma dilution at OD CoV-2 proteins. Overall, AT1R-Ab levels were not significantly different between controls and cases with either identify any significant associations, the trends we observed are consistent with a recent report of an association between AT1R-Ab and worse outcomes in patients hospitalized for COVID-19, 1 7 5 suggesting they may play a role in endothelial activation during COVID-19 (14). Amongst severe COVID-19 cases, where HLA typing was available, half of AT1R-Ab positive or have had pre-existing AT1R-Ab not associated with their SARS-CoV-2 infection. Alternatively, CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted July 5, 2021. ; https://doi.org/10.1101/2021.06.30.21259796 doi: medRxiv preprint

COVID-19 update: Covid-19-associated coagulopathy

We identified a region of potential structural homology between AT1R and the S2 domain of 1 8 4 SARS-CoV-2 Spike; however, among controls without SARS-CoV-2 there was no clear 1 8 5 association between AT1R-Ab reactivity and low-level reactivity against SARS-CoV-2 Spike 1 8 6 trimer arguing against cross-reactivity as a driver of AT1R-Ab development. In addition, some 1 8 7 controls reacted against RBD and NP, suggesting that the reactivity we detected may represent likely that these participants had experienced an unrecognized SARS-CoV-2 infection given Our study was limited by small sample size, particularly amongst our severe COVID-19 cases. In addition, as we did not have samples from cases that pre-dated their SARS-CoV-2 infections 1 9 4 we could not determine whether AT1R-Ab were pre-existing or developed in the setting of An improved understanding of the mechanisms driving COVID-19-associated coagulopathy 1 9 8 could point to therapies to lessen COVID-19 morbidity and mortality. We did not identify a 1 9 9 significant association between AT1R-Ab and COVID-19 severity in this small case-control 2 0 0 study, but the trends we observed support a possible association between AT1R-Ab and . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. Foundation [to JDG]. The funders had no role in study design, data collection and analysis, . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. References: 2 1 7 It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted July 5, 2021. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted July 5, 2021. ; https://doi.org/10.1101/2021.06.30.21259796 doi: medRxiv preprint AT1R-Ab negative AT1R-Ab positive COVID-19 case