Metformin, a first-line drug for type-2 diabetes, has been shown to improve locomotor recovery after spinal cord injury. However, there are studies reporting no beneficial effect. Recently, we found that high dose of metformin (200 mg/kg, intraperitoneal) and acute phase administration (immediately after injury) led to increased mortality and limited locomotor function recovery. Consequently, we used a lower dose (100 mg/kg, i.p.) metformin in mice, and compared the effect of immediate administration after spinal cord injury (acute phase) with that of administration at 3 days post-injury (subacute phase). Our data showed that metformin treatment starting at the subacute phase significantly improved mouse locomotor function evaluated by Basso Mouse Scale (BMS) scoring. Immunohistochemical studies also revealed significant inhibitions of microglia/macrophage activation and astrogliosis at the lesion site. Furthermore, metformin treatment at the subacute phase reduced neutrophil infiltration. These changes were in parallel with the increased survival rate of spinal neurons in animals treated with metformin. These findings suggest that low-dose metformin treatment for subacute spinal cord injury can effectively improve the functional recovery possibly through anti-inflammation and neuroprotection. This study was approved by the Institute Animal Care and Use Committee at the University of Texas Medical Branch (approval No. 1008041C) in 2010.