Methylprednisolone Attenuates Aquaporin-4Antibody-induced Cytotoxicity In Cultured Rat Cortical Astrocytes And Neurons

Objective: Neuromyelitis optica (NMO) is a severe inflammatorydemyelinating disease of the central nervous system(CNS) that predominantlyaffects the optic nerve and spinal cord. Over the last decade, there has been agrowing debate on whether NMO is a distinct entity or a form of MultipleSclerosis (MS). In2004, a highly specific serum autoantibody (NMO-IgG)was discovered in patients with NMO, and, subsequently, the aquaporin-4(AQP4), the most abundant water channel in the brain, was identified as thetarget antigen of NMO-IgG. A series of evidence indicate that AQP4antibodynot only as a disease-specific marker, which helped to distinguish NMO fromMS, but also plays a pivotal role in the pathogenesis of NMO, and NMO is anantibody-mediated humoral autoimmune diseases. The sensitivity andspecificity of the AQP4antibody in clinical diagnosis of NMO were73%and91%respectively; and AQP4antibody titers correlate well with the clinicalattacks and severity of NMO and decrease after treatment ofimmunosuppressive agents. The immunohistological studies revealed the lossof AQP4and glial fibrillary acidic protein with relatively preserved myelinbasic protein in the active lesions of NMO. In in vitro models, AQP4antibodyspecifically binding to human astrocytes induces reversible AQP4internalization, and increases permeability of blood-brain barrier. AQP4antibody also can impair sodium-dependent glutamate transport and activatecomplement cascade leading to complement-dependent cytotoxicity. However,these results came from purified astrocyte cultures or AQP4transfected cellsculture, the cytotoxicity of AQP4antibody in mixed cortical cultures has notbeen studied. In addition, the low-dose corticosteroid monotherapy is effectivein reducing relapse in NMO. The beneficial effect of corticosteroid areprobably mediated by immunosuppression, anti-inflammatory, and reducing cell edema. In the present study, we first investigated the effect of AQP4antibody-positive sera of different titers on astrocytes and neurons in primarycultured rat cortical cells, then tested whether methylprednisolone can reduceAQP4antibody-induced cytotoxicity.Methods:1Primary cultured cortical cells were prepared using Wister neonatal rats andexposed to the control sera, the sera of AQP4antibody titers1:10,1:320and1:1000for6hours, the morphology and quantity of astrocytes and neuronswere observed by immunofluorescence staining.2Primary cultured cortical cells were obtained from Wister neonatal rats andwere randomly divided into the control group, the AQP4antibody titer1:1000group, the alone methylprednisolone group (0.8umol), and themethylprednisolone(0.8umol)+AQP4antibody titer1:1000group. The fourgroups were added to the corresponding serum and/or methylprednisolone.After6hours of treatment, the numbers and morphological changes ofastrocytes and neurons were observed by immunofluorescence staining.Results:1Under the fluorescence microscope, astrocytes and neurons had clear cellbodys, thick and long processes in control group. At all in AQP4antibodypositive groups, the majority of astrocytes exhibited cellular swelling andhypertrophy, accompanied by the processes reduced, shorten andfragmentation. The neuronal processes also revealed the similar changes. Incontrast, the methylprednisolone pretreatment group could significantly reducethese morphological changes. Methylprednisolone alone had no significantinfluence on cellular morphology.2The exposure of primary cultured cortical cells to the AQP4antibodypositive sera induced significant decreased of the percentage of astrocytes andneurons as compared with the control group. The percentage of astrocyteswere52.22±3.75%,45.83±4.82%,35.77±5.91%and24.68±3.40%in thecontrol group, AQP4antibody title1:10,1:320and1:1000group respectively.The percentage of neurons were44.93±2.39%in the control group,39.00± 4.33%,32.21±2.67%and23.35±2.65%in1:10,1:320and1:1000grouprespectively. One way ANOVA analysis showed statistical significantdifferences among the four groups (F=48.49, P＜0.05for astrocytes andF=71.41, P＜0.05for neurons respectively).3The primary cultured cortical cells were pretreatment withmethylprednisolone group0.8umol for4hours before they were exposed to1:1000titer AQP4antibody positive sera. The percentage of astrocytes andneurons(44.52±3.71%and42.11±5.94%respectively) were significantlyincreased as compared with those of the AQP4antibody titer1:1000group(24.68±3.40%and23.35±2.65%respectively)(p<0.05). We aslo findmethylprednisolone pretreatment alone caused a reduction ofastrocytes(43.29±3.80%and52.22±3.75%respectively, P=0.00) while had noinfluence on the percentage of neurons(43.10±5.56%and44.93±2.39%respectively，P=0.42).Conclusions:1AQP4antibody induces titer-dependent cytotoxicity of astrocytes andneurons in primary cultured rat cortical cells.2Methylprednisolone pretreatment can reduce AQP4antibody-inducedcytotoxicity in cultured rat cortical cells.3The present study supports the AQP4antibody involved in NMOpathogenesis, and also provides theoretical evidence for the clinical detectionof AQP4antibody and hormone treatment for NMO.