In the past 10 years, neuromyelitis optica (NMO) has evolved from Devic's categorical clinical description into a broader disease spectrum. Serum IgG antibodies have been identified in NMO patients with the water channel aquaporin-4 (AQP4) as their main target antigen. AQP4 antibodies/NMO-IgG have been shown to be a highly specific and moderately sensitive serum biomarker for NMO. The immunopathology of NMO lesions supports that anti-AQP4 antibodies/NMO-IgG are involved in the pathogenesis of NMO. In vitro studies have demonstrated that human NMO-IgG induce necrosis and impair glutamate transport in astrocytes. Certain ethnic groups, notably of Asian and African origin, seem to be more susceptible to NMO than others. The genetic background for these putative differences is not known, a weak human leucocyte antigen association has been identified. AQP4 gene variants could represent a genetic susceptibility factor for different clinical phenotypes within the NMO spectrum. Experimental models have been described including a double-transgenic myelin-specific B- and T-cell mouse. NMO-like disease has been induced with passive transfer of human anti-AQP4 antibodies to the plasma of mice with pre-established experimental autoimmune encephalomyelitis or by intrathecal administration to naive mice. NMO may be characterized as a channelopathy of the central nervous system with autoimmune characteristics.