NEW YORK – Yale University researchers have developed a platform for identifying autoantibodies to extracellular proteins that could improve understanding of the role of autoantibodies across a variety of diseases.
Aaron Ring, assistant professor of immunobiology at Yale and lead developer of the platform, has also launched a company, Seranova Bio, that aims to use the technology for drug research and development.
Called REAP (for rapid extracellular antigen profiling), the approach screens patients samples against genetically barcoded human extracellular proteins expressed via a yeast-display system. Yeast cells that bind to autoantibodies in the patient samples are then isolated and the bound antigens are read out by sequencing the barcodes.
Ring noted that while autoantibodies are commonly associated with autoimmune disorders, there is substantial evidence that they also play a broader role within the body, including protecting against disease.
"There's a growing consensus that not all autoantibodies are bad and that some can be protective," he said, citing the example of breast cancer patients who have been found to produce anti-HER2 antibodies. "They are basically making their own trastuzumab, and they tend to live longer as a result."
Similarly, patients with autoimmune disorders who make autoantibodies against immune system components like cytokines have been found to have less severe disease. Biogen's Alzheimer's drug aducanumab is based on an autoantibody that attacks amyloid beta plaques that was discovered in elderly individuals who showed little evidence of cognitive decline, Ring said.
"The point is that we know the immune system can generate these rare but really interesting responses that can have dramatic impacts on health outcomes," he said. "And understanding those immune responses can give us insights into these diseases."
Researchers have developed tools for screening patient autoantibody repertoires using approaches like peptide and protein arrays. However, Ring said, these technologies have not been particularly good at detecting autoantibodies specific to extracellular proteins.
Extracellular proteins are commonly present in particular conformations and with particular post-translational modifications that existing platforms have had difficulty accounting for, he said.
"Most antibodies recognize their target in its native, folded form," he said. "So technologies where you chop up the protein into bits like phage approaches or peptide arrays are unfortunately not going to capture these really important antibodies."
To address this problem, Ring and his colleagues looked to yeast display technologies developed by protein engineers for biopharma and other applications.
"Yeast are surprisingly good at making human extracellular and secreted proteins," he said. "They have a lot of the same machinery to secrete proteins. They can perform post-translational modifications. So we made a big bet that if we could curate a very large library of human extracellular and secreted proteins and display them using yeast surface display, that we would capture a big chunk of the human exoproteome in that way."
In a BioRxiv preprint published in March, Ring and his colleagues used the REAP platform to screen blood samples from autoimmune polyendocrinopathy candidiasis ectodermal dystrophy (APECED) patients and systemic lupus erythematosus (SLE) patients against 2,688 human extracellular proteins. They identified a number of known autoantibody reactivities in APECED as well as a number of previously unidentified autoantibodies. In the SLE patients, they identified a range of new autoantibody reactivities including several correlated with disease severity and specific forms of the condition.
Earlier this month, Ring and colleagues published a study in Nature using the REAP technology to characterize the autoantibody repertoire in COVID-19 patients. Screening samples from 194 SARS-CoV-2 infected patients and healthcare workers against 2,770 proteins, the researchers found that COVID-19 patients had significantly higher levels of autoantibody reactivity compared to uninfected controls and that these autoantibodies targeted a wide range of tissues and biological functions.
They found that autoantibody profiles showed correlation with specific clinical characteristics in patients and disease severity. They also observed that many of the tissue-specific autoantibodies they identified were present in tissues that have been implicated in post-COVID-19 syndrome, suggesting a potential role for autoantibodies in this phenomenon.
"At a global level, it was shocking to see just how high the level of autoimmunity was in these patients, even in patients who had mild or asymptomatic disease," Ring said, though he noted that other relatively common infections like influenza might exhibit similar levels of autoimmunity.
He and his coauthors have continued to monitor some of the patients from the original study throughout the year following the initial sampling and have found that in some, the autoantibodies generated as part of the immune response to the virus have persisted.
His team is now collaborating with his Yale colleague Akiko Iwasaki, coauthor on the Nature study, to explore whether autoantibodies could be involved in the long-COVID condition reported in some patients.
"There could be multiple etiologies of long COVID," Ring said. "It doesn't have to be just one thing, but this is a very plausible mechanism."
Ring's lab has licensed the REAP technology to Seranova Bio, the New Haven, Connecticut-based firm he launched roughly a month ago with seed funding from Foresite Capital. He said that the company aims to create a "serological atlas," which it has named SeraVista, by running the REAP platform on samples from tens of thousands of patients and healthy individuals and correlating their autoantibodies profiles to health outcomes and phenotypes.
The company plans to use this data to learn what antibody responses have significant impacts on health outcomes and what their key targets are, information that Ring said it will use to inform its own drug development programs.
"Our mission in the first couple of years is to developed this atlas, SeraVista, so we can identify what the most compelling antibody responses are, and then to make drugs that model those responses in patients," he said.