Neuronal autoantibodies in the cerebrospinal fluid of 148 patients with schizophrenia and 151 healthy controls.

ABSTRACT

Schizophrenia is a syndrome with multiple etiologies, one of which is the potential for an autoimmune disease of the brain such as N-methyl-d-aspartate receptor (NMDAR) encephalitis, which can induce psychosis resembling schizophrenia. Here, we examined anti-neuronal autoantibodies related to psychosis using both cell- (CBA) and tissue-based assays (TBA) in the cerebrospinal fluid (CSF) of patients with chronic schizophrenia and control participants. First, we screened for the antibodies against leucine-rich glioma-inactivated 1 (LGI1), γ-aminobutyric acid B receptor (GABABR), dipeptidyl aminopeptidase-like protein 6 (DPPX), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR1/R2), and contactin-associated protein-like 2 (CASPR2) in 148 patients with schizophrenia. No antibodies were detected. Next, we performed CBA for NMDAR antibodies in 148 patients with schizophrenia and 151 age- and sex-matched controls. Although we detected relatively weak immunoreactivity for NMDAR in the CSFs of two patients with schizophrenia and three controls, no samples were positive when strict criteria were applied. For TBA in the rat hippocampus and cerebellum, we detected positive signals in the CSFs of 13 patients with schizophrenia and eight controls. Positive samples were analyzed for paraneoplastic syndrome and antinuclear antibodies using immunoblotting. The CSFs of nine patients and six controls were positive for dense fine speckle 70 (DFS70) antibodies. Additionally, antibodies against centromere protein (CENP)-A and CENP-B were detected in patients with schizophrenia. Our results suggest that autoantibodies against NMDAR, LG1, GABABR, DPPX, AMPAR1/R2, and CASPR2 are not associated with the pathogenesis of chronic schizophrenia. Moreover, we emphasize the importance of considering the effect of anti-DFS70 antibodies when analyzing autoantibodies in CSF samples. Conclusively, we obtained no evidence suggesting that the most frequent neuronal autoantibodies in the CSF play a role in the pathogenesis of schizophrenia, even in our sample.