IgLON5 belongs to the IgLON family of cell adhesion molecules, a group of proteins that sit on the outer membrane of neurons and help nerve cells recognize one another, form connections, and maintain stable circuits. These proteins are heavily modified with sugar chains and are expressed prominently in brain regions that govern sleep and breathing, including the hypothalamus and brainstem. When antibodies bind IgLON5, patients can develop what clinicians now call IgLON5 antibody-associated autoimmune encephalitis, a syndrome first described in the last decade that blends features of neurodegeneration with autoimmunity. Unlike many autoimmune encephalitides that respond briskly to immunotherapy, IgLON5 disease often follows a chronic, relapsing course and can be fatal when breathing during sleep goes untreated.

The clinical picture of IgLON5 disease is distinctive. Patients frequently suffer from severe, sometimes life-threatening sleep-disordered breathing, with abnormal sleep architecture, vocalizations, and stridor, a high-pitched wheezing caused by obstructed airflow. Movement disorders, swallowing difficulties, and cognitive decline may follow. Many patients carry certain immune system genes that increase susceptibility, and autopsy studies have revealed unusual deposits of tau protein in the hypothalamus and brainstem, linking this syndrome to both inflammatory and neurodegenerative mechanisms. Because the symptoms overlap with more common neurological conditions, diagnosis depends on detecting IgLON5 antibodies in blood and cerebrospinal fluid, a test that has only recently become widely available.

What has remained unclear is exactly where on the IgLON5 protein the pathogenic antibodies attach, and whether different patients produce antibodies that recognize the same molecular surface or a variety of epitopes. Answering that question matters for several reasons. The location of the antibody binding site can determine whether antibodies simply flag neurons for destruction by the complement system, whether they interfere with the protein’s adhesive function, or whether they cross-link IgLON5 molecules and disrupt signaling. Knowing the epitope also opens the door to rational diagnostic assays, patient stratification, and potentially therapies designed to block the antibody-protein interaction directly.

To resolve these questions, the research team turned to structural biology, combining X-ray crystallography and cryo-electron microscopy with binding studies using antibodies derived from affected patients. The IgLON5 protein has an immunoglobulin-like fold, arranged in domains that project from the neuronal membrane, and the researchers determined the three-dimensional structure of these domains both alone and in complex with antibody fragments. This approach allowed them to visualize, atom by atom, the surfaces of IgLON5 that patient antibodies engage, and to compare how different antibodies approach the same target.

The structures revealed that patient autoantibodies converge on defined surfaces of the IgLON5 molecule, with the binding interface shaped by a combination of protein side chains and, importantly, the sugar modifications that decorate the protein. IgLON family proteins are among the most heavily glycosylated molecules in the brain, and the study indicates that these carbohydrate structures are not merely decorative; they form part of the landscape that antibodies recognize. This finding helps explain why IgLON5 is immunogenic in some individuals and why antibody detection assays can vary in sensitivity depending on how the protein is produced and modified in the laboratory.

Beyond mapping the epitopes, the work addressed how antibody binding might translate into neuronal injury. Autoantibodies in autoimmune encephalitis can act through several mechanisms: they may cross-link target proteins and cause their internalization, they may block protein-protein interactions essential for synaptic stability, or they may activate complement and recruit immune cells that destroy the neuron. The structural data suggest that the antibodies bind bivalent, bridging IgLON5 molecules on the neuronal surface, a geometry consistent with clustering of the protein and with downstream effector activation. Such clustering could perturb the adhesive contacts that IgLON5 normally maintains between neurons, contributing to the circuit dysfunction seen in patients.

The findings also carry implications for understanding the broader family of IgLON proteins. Because IgLON5 shares structural features with its relatives, including IgLON1, IgLON2, and IgLON3, the mapped antibody-binding surfaces provide a template for investigating whether related autoantibodies, reported in a small number of patients, engage similar or distinct regions. More broadly, the study adds to a growing body of structural work on neural autoantigens, including the NMDA receptor, LGI1, and DPPX, showing that each autoimmune encephalitis syndrome has its own molecular logic. In IgLON5 disease, that logic appears to involve recognition of a highly glycosylated adhesion molecule in brain regions with limited regenerative capacity.

For patients, the practical significance of the study lies in improved diagnostics and, eventually, targeted treatment. Current therapy for IgLON5 disease relies on immunoglobulins, plasma exchange, and immunosuppressants, approaches that blunt the immune attack broadly but do not specifically prevent antibodies from binding their target. A structural map of the epitope makes it conceivable to design decoy molecules that soak up pathogenic antibodies, or to engineer diagnostic antigens that better reflect the native, glycosylated form of IgLON5 found on neurons. It also enables the monitoring of antibody specificity over the course of disease, which could help clinicians predict relapses and tailor immunotherapy intensity.

The research also underscores a recurring theme in modern neuroimmunology: the same molecular target can drive both inflammation and degeneration. In IgLON5 disease, antibody binding to a neuronal adhesion protein may initiate immune-mediated injury while simultaneously destabilizing the protein complexes that keep neurons healthy over decades, promoting the tau pathology observed at autopsy. Disentangling these processes is one of the field’s central challenges, and structural knowledge of the antibody-antigen interface is a critical step. By showing precisely how the immune system reads the surface of IgLON5, the study transforms a mysterious clinical syndrome into a mechanistically defined disease, and provides the molecular blueprint from which new diagnostics and therapies can be built.

Subject of Research: Structural basis of IgLON5 autoantibody recognition in autoimmune encephalitis

Article Title: Structural basis of IgLON5 autoantibody recognition in autoimmune encephalitis

Article References: Roux, A., Schelling, R., Vinyals-Sales, D., Sabater, L., Winiger, R. R., Senyuz, I., Lin, A., Mathias, A., Du Pasquier, R., Gaig, C., Dalmau, J., Foglierini, M., & Perez, L. (2026). Structural basis of IgLON5 autoantibody recognition in autoimmune encephalitis. Nature Communications. https://doi.org/10.1038/s41467-026-77812-6

Image Credits: AI Generated

DOI: 10.1038/s41467-026-77812-6

Keywords: IgLON5, autoimmune encephalitis, autoantibodies, structural biology, neuroimmunology, sleep-disordered breathing, neuronal adhesion proteins, glycosylation, epitope mapping, tau pathology, immunotherapy, Nature Communications