Autoantibodies and inflammatory reactions to COVID-19: we explore the work of Dr Paul Utz at Stanford University School of Medicine
Biotech writer Olivia Siviter explores research by Dr Paul Utz at Stanford University School of Medicine, presented at the Antibody Engineering & Therapeutics 2021 conference.
In the pandemic, rheumatologists were seeing people develop new autoimmune diseases as a result of contracting COVID-19.
Previous studies of critically ill COVID-19 patients suggested that due to an increased concentration of autoantibodies, Covid-19 was either causing development of new autoimmune diseases or patients had underlying autoimmune diseases that had not yet been detected.
However, there had yet to be a differentiation between which of the two was actually causing severe reactions to COVID-19.
To investigate this, blood serum was taken from patients with COVID-19 and added it to a mouse model with a damaged endothelium.
Research from Dr Jason Knight, University of Michigan Health found that patients with more severe COVID-19 formed much larger clots at the site of the damaged endothelium that mimicked an inflammatory response which would be similar to that seen in patients with autoimmune diseases such as cryopyrin-associated periodic syndrome (CAPS).
Serum from patients with a less severe reaction and a control group showed nothing more than a normal inflammatory response to the damaged endothelium suggesting that patients with a severe reaction to COVID-19 were having an autoimmune reaction, most likely caused by autoantibodies. To further characterise the autoimmune response, the blood clots were eluted, to look for myeloperoxidase and DNA, the components of neutrophil extracellular traps (NETs).
NETs are released as a result of neutrophils undergoing netosis as an immune response and are implicated in many autoimmune diseases. NETs were present suggesting that the blood clots were formed because of the presence of autoantibodies.
These findings led Dr Paul Utz’s group at Stanford University School of Medicine to ask whether there are specific autoantigens present in COVID-19 and whether the inflammatory reaction is a result of pre-existing or new autoantibodies?
The lab took samples from patients with SARS-CoV-2 at beginning of pandemic and performed anti-nuclear antibody tests. They found autoantibodies in a quarter of the patients.
Interestingly, the patterns that appeared in the assays were unique, some were homogeneous, some were speckled, and some had a nucleoid pattern which suggest heterogeneity in the responses.
They also looked for common autoantibodies against dsDNA and ANCA antibodies (anti-neutrophil cytoplasmic antibody) - almost no patients had these.
To further understand the autoantibodies in patients with severe COVID-19, 3 different arrays were created using a Luminex bead-based assay; an autoantigen array that had typical autoantigens from connective tissue diseases, a cytokine array to look at secreted proteins, and a COVID-19 viral array that had many proteins from COVID-19 and other viruses on it.
They found that 50% of COVID-19 patents had autoantibodies compared to about 10-15% in healthy controls. They identified antibodies that are typically found in patients with autoimmune myositis, specifically against the translational apparatus: signal recognition particle and tRNA synthetases.
They also found antibodies against scleroderma antigens, in some cases at the levels you would see in patients that have scleroderma.
Subsequent research on COVID-19 patients found pathogenic antibodies against Beta 2 glycoprotein, C1q (involved in binding to immune complexes and removing them from an inflammatory environment) and ACE2.
Logically, it makes sense to think that autoantibodies against ACE2 could be protective, due to it being the binding site of the SARS-CoV-2 spike protein.
However this cannot be the case, as samples were taken from hospitalised patients. It was suggested that maybe ACE2 was upregulated in response and that compliment deposition activates more inflammatory cascades.
A paper by Dr Paul Bastard et al, published before their second phase of research looked at over 1000 severe COVID-19 cases and found that around 12% of men and 3% of women have antibodies against interferon alpha.
Interestingly, no patients with mild disease (0/663) had these autoantibodies. ACAs block binding to the receptor and are likely to be pathogenic.
From this research, they proposed that as most of us do not have these antibodies, the interferon is upregulated when we are infected to fight the virus but if you have blocking antibodies the virus replicates much better, causing severe disease.
The paper also suggested that nearly as many as 20% of people over the age of 70 have blocking anti-interferon antibodies.
As a result of these findings, Dr Utz’s lab carried out another study of severe COVID-19 patients compared to patients with APS1.
They found that patients with severe COVID-19 had a high concentration of ACAs and that these autoantibodies must have been induced, as they were absent at the beginning of the longitudinal study but present towards the end.
One patient even had no antibodies at day 22 but a high concentration 29, suggesting that you can develop these antibodies during the hospital course.
The key take-home point from Dr Utz’s presentation is that some infectious diseases are fundamentally not infectious diseases, they are actually autoimmune diseases.
