Recently, the disease-specific antibody was found in the sera from neuromyelitis optica (NMO) patients, and its target was identified as aquaporin-4 (AQP4), mainly expressed in astroglial foot processes. In our immunohistochemical studies, loss of AQP4 and glial fibrillary acidic protein (GFAP) was evident in about 90% of NMO lesions, especially in perivascular areas of acute inflammatory lesions where immunoglobulins and complements were deposited. In contrast, myelin basic protein (MBP)-stained myelinated fibers were relatively preserved in those lesions, which probably suggested the secondary damage of myelin sheaths following astrocytic damage. Recently, there are developing evidences of the effect of AQP4 antibody in vitro or in vivo. In HEK293 cells transfected with AQP4, AQP4 antibody could bind to the membrane AQP4, and induced the degradation and endocytosis of AQP4 in complement-dependent manner. In vitro experiments by primary cultured astrocytes, AQP4 antibody had cytotoxic effects with complement, and also could impair the astrocytic function such as the maintenance of the blood brain barrier or glutamate homeostasis. In vivo study, the lesions lacking AQP4 and GFAP was appeared by passive-transferred Lewis rats with human purified IgG from NMO patients. Furthermore, in cerebrospinal fluid (CSF) biomarker study, astrocytic damage reflected by marked increase of CSF-GFAP, far severe than demyelination (CSF-MBP), was evident in NMO but not in classical multiple sclerosis (MS). These evidences suggested the pathogenic role of AQP4 antibody with astrocytopathy in NMO. Now it is indispensable to check the AQP4 antibody,and is important to reconsider the role of astrocyte in demyelinating disorders.