We assessed whether chronic treadmill exercise attenuated the depressive phenotype induced by restraint stress in ovariectomized mice (OVX). Immobility of OVX in the forced swimming test was comparable to that of sham mice (CON) regardless of the postoperative time. Immobility was also no difference between restrained mice (exposure to periodic restraint for 21 days; RST) and control mice (CON) on post-exposure 2nd and 9th day, but not 15th day. In contrast, the immobility of ovariectomized mice with repeated stress (OVX + RST) was profoundly enhanced compared to ovariectomized mice-alone (OVX), and this effect was reversed by chronic exercise (19 m/min, 60 min/day, 5 days/week for 8 weeks; OVX + RST + Ex) or fluoxetine administration (20 mg/kg, OVX + RST + Flu). In parallel with behavioral data, the immunoreactivity of Ki-67 and doublecortin (DCX) in OVX was significantly decreased by repeated stress. However, the reduced numbers of Ki-67- and DCX-positive cells in OVX + RST were restored in response to chronic exercise (OVX + RST + Ex) and fluoxetine (OVX + RST + Flu). In addition, the expression pattern of cAMP response element-binding protein (CREB) and calcium-calmodulin-dependent kinase IV (CaMKIV) was similar to that of the hippocampal proliferation and neurogenesis markers (Ki-67 and DCX, respectively). These results suggest that menopausal depression may be induced by an interaction between repeated stress and low hormone levels, rather than a deficit in ovarian secretion alone, which can be improved by chronic exercise.