Valproic acid (VPA) is widely used to treat epilepsy, bipolar disorders and migraine. Clinical studies showed that dosing VPA during pregnancy causes an increase in the risk of autism spectrum disorders (ASD) and mental retardation in children. Thus, rodents prenatally exposed to VPA are considered as animal models of ASD. We have also demonstrated that mice prenatally exposed to VPA at embryonic day 12.5 (E12.5) display ASD-like behaviors, such as social interaction deficit, anxiety-like behavior and memory deficit. Several clinical and animal studies suggested that dysfunction of glutamatergic and GABAergic systems are involved in the pathology of ASD. In the circumstances, the present study examined whether prenatal exposure to VPA affects dopaminergic systems in mice. VPA (500 mg/kg, i.p.) or saline was injected into pregnant mice on gestational days 12.5, and behavioral and neurochemical analyses were carried out in the offspring at 8 weeks old. Prenatal exposure to VPA at E12.5 decreased methamphetamine (METH)-induced hyperlocomotion. In vivo microdialysis study demonstrated that the prenatal VPA exposure attenuated METH-induced increase in extracellular dopamine (DA) level, without affecting noradrenaline and serotonin levels, in the prefrontal cortex (PFC). In addition, the prenatal VPA exposure decreased METH-induced c-Fos expression in the cells of the PFC. Furthermore, the prenatal VPA exposure decreased mRNA levels of DA D1, D2 receptors, calbindin-D28k and dysbindin. The VPA-induced changes in these parameters were not observed in the striatum. Taken together with the evidence that calbindin-D28k and dysbindin regulate DA release, the present finding suggests that the prenatal VPA exposure causes dysfunction of dopaminergic system in the PFC, resulting in a decrease in METH sensitivity. The study also implies that dopaminergic system may be involved in abnormal behaviors in VPA model mice.