To establish semi-quantitative high sensitivity ELISA for detecting anti-NMDA receptor (NMDAR) autoantibodies (autoAbs) from patients with anti-NMDAR encephalitis (ANRE).

AutoAbs targeting the N-terminal domain of the NR1 subunit (NR1-NTD) of the NMDA receptor (NMDAR) cause ANRE which is the most prevalent form of autoimmune encephalitis. A fixed cell based assay (fixed-CBA) using HEK293 cells expressing NMDAR NR1 protein is widely used for detecting anti-NMDAR IgG autoAbs. Although the fixed-CBA is well accepted in the field, it is less-sensitive in serum samples due to relatively high background staining and gives only antibody titers. Although ELISA is a standard assay to detect antibodies in general, anti-NMDAR autoAb detection has not been successfully demonstrated using regular peptide or protein ELISA.

A novel ELISA was developed using NR1-NTD protein (19-403 aa) with a C-terminal biotinylated Avi-tag as the antigen used for immobilization on streptavidin coated plates. ELISA sensitivity and specificity were assessed using monoclonal antibodies from ANRE patients (Ly et al. 2018). Serum and CSF samples from ANRE patients at the University of Texas Southwestern in both acute and recovery phase were tested by NR1-NTD ELISA and the results were compared with previous data obtained by clinically available fixed-CBA. 

After observing poor sensitivity with a typical random NTD-NR1 protein coating ELISA method, a unidirectional coating method was established to detect weak and low abundance mAbs from ANRE patients. Importantly, this ELISA detected IgG signals in serum samples that were originally judged negative by clinically acquired fixed-CBA from recovering ANRE patients who still had residual symptoms at blood sampling. The pseudo-concentrations of IgG were lower in these patients in recovery phase compared to acute phase.

Novel semi-quantitative high sensitivity NTD-NR1 ELISA may enable us to detect anti-NMDAR IgG autoAbs in serum samples from recovering ANRE patients who still suffer from residual anti-NMDAR autoAbs.