A new study reveals a critical vulnerability in a brain receptor targeted by a rare autoimmune disease.
Researchers have moved a step closer to new treatments for a rare and often devastating autoimmune disease that interferes with how the brain communicates.
The study highlights a vulnerable point in the disease process, one that could eventually be used to design more precise therapies. The condition develops when the immune system mistakenly targets the NMDA receptor, a protein essential for learning, memory, and normal brain signaling. By identifying where this attack occurs, the work also opens the possibility of a future blood test that could flag the disease earlier, when current treatments may be more effective.
The research was conducted at Oregon Health & Science University and published in the journal Science Advances.
The disorder gained public attention through the bestselling memoir and the 2016 film “Brain on Fire,” but it remains uncommon. Each year, it affects an estimated 1 in a million people, most frequently young adults in their 20s and 30s. Despite its rarity, the illness can progress rapidly and cause life-threatening neurological symptoms.
In people with the condition, the immune system produces anti-NMDA receptor autoantibodies that attach to NMDA receptors in the brain and disrupt their function. This immune-driven damage can lead to dramatic changes in behavior and cognition, severe memory loss, seizures, and, in some cases, death.
Pinpointing a Molecular Target
In the new study, scientists focused on identifying exactly where these harmful antibodies bind. They discovered specific regions on a subunit of the NMDA receptor that appear to be central to the disease process. Blocking these regions, the researchers suggest, could help stop or even reverse the progression of symptoms.
Lead author Junhoe Kim, Ph.D., a postdoctoral fellow in the OHSU Vollum Institute, analyzed a complete set of anti-NMDA receptor autoantibodies taken from a mouse model that OHSU researchers had previously developed to closely mimic the human disease. He then compared these results with high-resolution images of autoantibodies isolated from patients.
The analysis revealed a striking overlap. The antibody binding sites identified in the mouse model were located in the same regions of the NMDA receptor as those found in people with the disorder, strengthening the case that the model accurately reflects the human condition and that the identified sites are promising targets for future therapies.
“We have really solid evidence because the autoantibody binding sites that Junhoe identified overlap with those from people,” said senior author Eric Gouaux, Ph.D., senior scientist in the Vollum and an investigator with the Howard Hughes Medical Institute. “We’re focused now on this area as literally a hot spot for the interaction that underpins at least one component of the disease.”
Kim said researchers knew generally where to look.
“From previous studies, people knew where the antibodies might bind,” he said. “But we collected the entire native autoimmune antibody panel from a mouse model with the disease, and we elucidated where specifically they bind onto the receptor.”
Advanced Imaging Reveals Key Details
They made the discovery using near-atomic imaging at the Pacific Northwest Cryo-EM Center, housed at OHSU’s South Waterfront campus and one of three national centers for the state-of-the art imaging technology. It’s operated jointly by OHSU and the Pacific Northwest National Laboratory, and funded by the National Institutes of Health.
“Nearly all of the antibodies bound to a single domain of the receptor that happens to be the part of the receptor that’s simplest to target,” Gouaux said. “It’s a super exciting result, actually.”
Co-author Gary Westbrook, M.D., a neurologist and senior scientist in the Vollum Institute, said the discovery may clear the way for drug companies to develop a therapeutic agent that could specifically target the binding sites causing the disease. Current therapies involving immunosuppression don’t always work and patients can relapse, he said.
“More specific approaches are definitely needed,” he said.
Reference: “Cryo-EM of autoantibody-bound NMDA receptors reveals antigenic hotspots in an active immunization model of anti-NMDAR encephalitis” by Junhoe Kim, Farzad Jalali-Yazdi, Brian E. Jones, Gary L. Westbrook and Eric Gouaux, 14 January 2026, Science Advances.
DOI: 10.1126/sciadv.aeb4249
The research was supported by the National Research Foundation of Korea, award RS202400334731; the National Institute of Mental Health, and the National Institute of Neurological Disorders and Stroke, both of the National Institutes of Health, under award numbers F32MH115595, R01NS117371 and R01NS038631; the Howard Hughes Medical Institute and Jennifer and Bernard LaCroute. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Never miss a breakthrough: Join the SciTechDaily newsletter.
Follow us on Google and Google News.