An increased inflammatory response and deficient synthesis of neurotrophic factors (NTFs) may contribute to the etiology of depression. However, the interrelationship between inflammation and NTFs is unknown. Recently, ethyl-eicosapentaenoate (EPA) has been used to treat depression. The mechanism by which EPA benefits depression is also unclear. Using the olfactory bulbectomized (OB) rat model of depression, this study evaluated two pathways from bulbectomy to the induction of depression-like changes (the inflammation-hypothalamic-pituitary-adrenal axis-stress response pathway and inflammation-nerve growth factor-memory pathway) and the effect of EPA on these pathways. When compared with sham-operated rats fed a control diet, significantly increased locomotor and rearing activities in an "open field," impaired memory in the Morris water maze, increased expression of corticotrophin-releasing factor (CRF), and increased secretion of corticosterone were found in OB rats. mRNA expression of nerve growth factor (NGF) was significantly lower in the hippocampus, and phospholipase A2 (PLA2) was higher in the hypothalamus; this change was associated with increased interleukin-1beta (IL-1beta) and prostaglandin E2 (PGE2) in the serum and brain. EPA treatments normalized these behavioral impairments and reduced CRF expression and corticosterone secretion. EPA also reduced serum concentrations of IL-1beta and PGE2, but reversed NGF reduction. Similar to the effects of EPA, the anti-inflammatory drug celecoxib significantly reduced blood PGE2, IL-1beta, and corticosterone concentrations and increased NGF expression in OB rats. Furthermore, anti-NGF treatment blocked EPA effects on behavior. These results suggest that an interaction exists between inflammation and NGF in the depression model. EPA may improve depression via its anti-inflammation properties and the upregulation of NGF.