Memantine shows clinically relevant efficacy in patients with Alzheimer's disease and Parkinson's disease. Most in vivo and in vitro studies attribute the neuroprotective effects of memantine to the blockade of N-methyl-D-aspartate (NMDA) receptor on neurons. However, it cannot be excluded that mechanisms other than NMDA receptor blockade may contribute to the neuroprotective effects of this compound. To address this question, primary midbrain neuron–glia cultures and reconstituted cultures were used, and lipopolysaccharide (LPS), an endotoxin from bacteria, was used to produce inflammation-mediated dopaminergic (DA) neuronal death. Here, we show that memantine exerted both potent neurotrophic and neuroprotective effects on DA neurons in rat neuron–glia cultures. The neurotrophic effect of memantine was glia dependent, as memantine failed to show any positive effect on DA neurons in neuron-enriched cultures. More specifically, it seems to be that astroglia, not microglia, are the source of the memantine-elicited neurotrophic effects through the increased production of glial cell line-derived neurotrophic factor (GDNF). Mechanistic studies showed that GDNF upregulation was associated with histone hyperacetylation by inhibiting the cellular histone deacetylase activity. In addition, memantine also displays neuroprotective effects against LPS-induced DA neuronal damage through its inhibition of microglia activation showed by both OX-42 immunostaining and reduction of pro-inflammatory factor production, such as extracellular superoxide anion, intracellular reactive oxygen species, nitric oxide, prostaglandin E₂, and tumor necrosis factor-α. These results suggest that the neuroprotective effects of memantine shown in our cell culture studies are mediated in part through alternative novel mechanisms by reducing microglia-associated inflammation and by stimulating neurotrophic factor release from astroglia.

Memantine, a noncompetitive N-methyl-d-aspartate receptor antagonist with a mood-stabilizing effect, and an association between bipolar disorder and proinflammatory cytokine levels have been reported. Whether adding-on memantine would reduce cytokine levels and is more effective than valproic acid (VPA) alone in bipolar II disorder was investigated. A randomized, double-blind, controlled, 12-week study was conducted. Patients undergoing regular VPA treatments were randomly assigned to a group: VPA + memantine (5 mg/d) (n = 106) or VPA + placebo (n = 108). The Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS) were used to evaluate clinical response. Symptom severity, plasma tumor necrosis factor (TNF-), interleukin 6 (IL-6), IL-8, and IL-1 levels were examined during weeks 0, 1, 2, 4, 8, and 12. To adjust within-subject dependence over repeated assessments, multiple linear regressions with generalized estimating equation methods were used to examine the therapeutic effect. Tumor necrosis factor levels were significantly lower in the VPA + memantine group than in the VPA + placebo group (P = 0.013). Posttreatment HDRS and YMRS scores decreased significantly in both groups, but not significant, nor was the other between-group cytokine level difference pretreatment and posttreatment. The HDRS score changes were significantly associated with IL-6 (P = 0.012) and IL-1 (P = 0.005) level changes and changes in YMRS score changes with TNF- (P = 0.005) level changes. Treating bipolar II depression with VPA + memantine may improve the plasma TNF- level. However, adding-on memantine may not improve clinical symptoms or cytokine levels other than TNF-. Clinical symptoms may be correlated with certain cytokines.