Many individuals continue to experience symptoms of COVID-19 months after recovery. This condition is known as post-COVID syndrome or long COVID. Long COVID can manifest in various ways, ranging from severe fatigue and brain fog to muscle and joint pain, shortness of breath, sleep disturbances, and tachycardia. Despite extensive research, the underlying causes of these symptoms remain unclear.
One of the leading hypotheses involves autoimmunity. Studies have shown that patients with long COVID retain circulating autoantibodies – antibodies that mistakenly recognize the body’s own tissues and cells as foreign. These autoantibodies may interact with proteins of the immune and nervous systems, as well as with cellular receptors, disrupting their normal function.
Such autoantibodies may not merely accompany the disease but may actively contribute to symptom development. For example, some patients with post-COVID syndrome experienced symptom improvement after procedures designed to reduce autoantibody levels in the blood. Furthermore, experiments in mice have demonstrated that transfer of antibodies from patients with fibromyalgia induces pain-related behaviors.
Researchers at the Amsterdam University Medical Center hypothesized that autoantibodies may play a direct role in the development of long COVID in at least some patients. To test this hypothesis, they transferred IgG isolated from patients with long COVID into mice.
Long COVID Patients Exhibit Altered Interferon Levels and Elevated GFAP, a Marker of Astrocyte Activation
The researchers examined 34 patients with long COVID whose symptoms persisted for at least six months after SARS-CoV-2 infection and compared them with 15 individuals who had fully recovered from COVID-19. Common symptoms among long COVID patients included fatigue, post-exertional malaise, and pain-related symptoms.
Biomarker analysis showed that patients with long COVID had reduced levels of interferon gamma (IFN-γ). Some patients also exhibited elevated levels of GFAP, a marker of astrocyte activation, and NFL, a marker of potential neuronal injury. Most inflammatory and neurodegenerative markers, however, did not differ from those in the control group.
Despite the absence of pronounced group-wide differences, the researchers observed substantial interindividual variability in biomarker profiles. Based on levels of GFAP, NFL, and IFN-β, patients were divided into three subgroups:
- LC-1 – characterized by signs of potential central nervous system involvement (elevated GFAP and NFL);
- LC-2 – characterized by elevated IFN-β levels;
- LC-3 – characterized by low IFN-β
This approach enabled the identification of patients with potential central nervous system involvement or immune dysregulation.
Protein and autoantibody analyses indicated that different long COVID subgroups may be associated with pathology in different organs and tissues:
- LC-1 – primarily associated with the nervous system;
- LC-2 – primarily associated with muscles;
- LC-3 – primarily associated with metabolically active tissues.
Transfer of IgG From Long COVID Patients Induces Pain Hypersensitivity in Mice
To investigate the role of antibodies in long COVID, the researchers injected mice with IgG purified from the blood of long COVID patients and healthy donors. Despite the absence of systemic inflammation, IgG from long COVID patients increased sensitivity to both mechanical and thermal stimuli. This hypersensitivity persisted for at least 15 days after injection.
The most pronounced effects were observed following transfer of IgG from the LC-1 and LC-3 subgroups. However, the pattern of symptoms differed between groups: LC-3 antibodies induced hypersensitivity within the first day, whereas the effects of LC-1 antibodies developed gradually over three days.
These findings demonstrate that IgG from long COVID patients can independently induce sensory abnormalities in mice.
Transfer of IgG From Long COVID Patients Affects Locomotor Activity in Mice
IgG from the LC-2 subgroup reduced locomotor activity by a moderate amount. Within 24 hours after injection, mice traveled approximately 40% less distance than control animals, primarily due to increased immobility time, while coordination and balance remained unaffected. IgG from the LC-1 and LC-3 subgroups did not affect locomotor activity.
Long COVID Patients Harbor Autoantibodies Against Distinct Targets
Proteomic analysis identified 134 autoantibodies that were elevated in long COVID patients compared with healthy donors. Each previously identified subgroup exhibited a distinct autoantibody profile: LC-1 was associated primarily with skin proteins, LC-2 with neuronal and immune targets, and LC-3 with proteins involved in inflammation and pain signaling.
Following transfer into mice, these IgG antibodies accumulated in various tissues, including the spinal cord, dorsal root ganglia, heart, and skeletal muscle. Despite this accumulation, there were no signs of glial or immune activation.
Autoantibodies and Biomarkers of Long COVID Persist for at Least Two Years
A follow-up examination approximately 2 years later demonstrated that several characteristic biomarkers of long COVID remained elevated. Some patients continued to exhibit increased levels of IFN-β and GFAP. IFN-γ levels no longer differed between long COVID patients and healthy controls. Two markers of neural injury – GFAP and NFL – became relatively elevated in the LC-2 subgroup, reaching levels comparable to those observed in LC-1.
Transfer of IgG isolated from long COVID patients again induced mechanical hypersensitivity in mice. The most consistent effects persisted in the LC-1 and LC-3 subgroups; however, in the follow-up experiment, a similar effect was also observed in the LC-2 subgroup.
Several subgroup-associated autoantibodies retained high reactivity throughout the 2 years, both at the initial assessment and at follow-up.
Conclusion
Autoantibodies from patients with long COVID induce persistent pain hypersensitivity in mice, suggesting a potential causal role for autoantibodies in the development of long COVID. This effect was reproduced in follow-up experiments conducted two years after the initial patient assessment.
At the two-year follow-up, patients continued to exhibit elevated type I interferon activity and markers of neural injury. Autoantibody profiles specific to each patient subgroup remained stable for at least two years.
These findings support the concept that long COVID comprises multiple biological subtypes and open new possibilities for immunotherapeutic approaches. Such therapies may focus on removing or neutralizing pathogenic antibodies. Long-term strategies targeting autoreactive plasma cells and B cells may also represent promising therapeutic options.
Reference
Transfer of IgG from long COVID patients induces symptomology in mice
