Brain-reactive autoantibodies in patients with systemic lupus erythematosus

Neuropsychiatric Systemic Lupus Erythematosus (NPSLE) is an autoimmune disease characterized by the production of brain-reactive autoantibodies. These autoantibodies are thought to be one of the main mediators of this condition, however little is known about their cellular and molecular target. To get inside this matter, we enrolled a cohort of 209 subjects including patients affected by Systemic Lupus Erythematosus (SLE, n=69), NPSLE (drug-naïve, n=36), multiple sclerosis (MS, n=22) as well as 82 age and gender matched healthy subjects. Sera from these subjects were used for immunostaining on rat brain through fluorescence microscopy and TEM, in parallel with immunofluorescence on human brain sections and cells (SH-SY5Y) to eventually confirm the results obtained in rat. As to the brain autoantibody target, western blot (WB) using wild type and transfected HEK293 cells, which do or do not express the RBFOXP3 gene (encoding the NeuN protein). Furthermore, ELISA was programmed to reveal the possible presence of the known circulating autoantibodies in SLE (anti: -Ro/SSA, -La/SSB, -RNP, -Sm, and -Rib-P) and autoantibodies to the aminoacidic sequence DWEYS. On rat brain sections, we observed autoantibody positive reaction in several areas (cortex, hippocampus, cerebellum, and others) exclusively using sera from patients with NPSLE and SLE but with prevalence and higher titer among the NPSLE patients (77,1% vs. 56,5%, respectively and 1:2,000-8,000 vs. 1:120-1,000, respectively). In order to consolidate the results obtained in rats, negative and positive sera from SLE and NPLSE patients as well as healthy and MS controls were selected (n=6) to be tested in human hippocampus and cerebellum sections. Autoantibody reactivity was found only using those sera positively reacting on rat tissues. Through ELISA, anti: -Ro/SSA, -La/SSB, -RNP, -Sm, -Rib-P and -DWEYS were found using rat-immunostained-positive sera from both NPSLE and SLE patients. In both patients’ populations, the presence of one or more types of these autoantibodies was not higher than 31,6% while a percentage was negative for all the above autoantibodies (17,9% and 31,6%, in NPSLE and SLE respectively). Hence, the presence of these autoantibodies could confirm only partially the reactivity according to the immunostaining results. In the rat cerebellum, the autoantibodies stained a cytoplasmic area also in proximity of the nucleus (namely perinuclear/cytoplasmatic pattern) or inside the nucleus (namely intranuclear pattern), both revealed using either fluorescence microscopy or TEM. In order to identify the cell type recognized by the autoantibodies in the cerebellum, we performed double staining, mixing the sera from patients with the antibody to NeuN (known to mark the granule cells), as well as antibodies to VGlut1, VAChT or GAD-65, marking glutamatergic axon terminals, cholinergic and GABAergic neurons, respectively. Autoantibody staining was localized exclusively within NeuN containing cells. Autoantibodies labelled NeuN containing neurons also in many other areas of the brain. We performed WB experiments to reveal NeuN as possible autoepitope using either wild type and transfected HEK293 cells, which do not or do express the RBFOXP3 gene, respectively. The patient’s sera never recognized the bands corresponding to the NeuN protein, despite these bands were labeled by the anti-NeuN antibody using transfected cells, proving that NeuN was not the autoantibody target. In conclusion, all together the da-ta obtained from this study demonstrate that (1) SLE and NPSLE could possess a mix of different types of autoantibodies reacting to neuronal cells (2) the entire brain could be the target of these autoantibodies (3) a high prevalence of brain reactive autoantibodies is found among NPLSE patients and (4) autoantibodies could react to the nucleus and/or cytoplasm of brain neuron cells.