# Acute neuronal cell death and neuroinflammation per se do not trigger secondary autoimmune encephalitis in mice

## TL;DR

This study shows that acute brain cell death and inflammation in mice do not lead to autoimmune encephalitis, even though some autoantibodies were detected.

## Contribution

The study challenges the hypothesis that neuronal damage and neuroinflammation alone trigger autoimmune encephalitis.

## Key findings

- Acute neuronal death and neuroinflammation in mice did not result in autoimmune encephalitis symptoms.
- Autoantibodies against brain antigens were rarely detected in mice with encephalitis.
- Lymphocyte infiltration and microgliosis were present but did not progress into autoimmune responses.

## Abstract

Patients with virus encephalitis, such as herpes simplex encephalitis and Japanese encephalitis frequently relapse with autoimmune encephalitides associated with neural autoantibodies. It has been hypothesized that the infection-induced damage to the central nervous system results in shedding of neural autoantigens, their presentation to the peripheral immune system, and initiation of a secondary autoimmune encephalitis that targets these autoantigens. To test this hypothesis, we utilized a transgenic mouse model of virus-like but sterile encephalitis. After induction of acute neuronal death in the hippocampus, we monitored the mice for encephalitis-like symptoms for up to 10 months, evaluated the degree of neuroinflammation at several time points and screened their plasma for autoantibodies against 49 different autoimmune disease-associated brain autoantibodies. Throughout the study period, we did not detect any symptoms of severe autoimmune encephalitis, like hyperactivity, circling, seizures, lethargy. Evaluation of microglia numbers and morphology revealed pronounced microgliosis 1-week after initial encephalitis induction, which decreased over time. Scattered lymphocyte infiltration was present at all times in hippocampi of encephalitis mice, and did not increase over time. Perivascular cuffs were not detected. Infiltrating lymphocytes mainly consisted of CD8+ T cells. B cell infiltration was rare and did not differ from healthy control mice. High-parameter immunophenotyping of peripheral blood leukocytes did not reveal any changes associated with an autoimmune response. Testing all plasma samples (n = 30/group) at a dilution of 1:100 for autoantibodies against 49 neural autoantigens gave only two positive results, namely one healthy control with anti-CASPR2 autoantibodies (IgG) and one post-encephalitis mouse with anti-homer-3 autoantibodies (IgM). Overall, these findings suggest that acute neuronal cell death and neuroinflammation per se are not sufficient to trigger downstream autoimmune encephalitis relapses in mice.