Rodents exposed prenatally to valproic acid (VPA) show autism-related behavioral abnormalities. We recently found that prenatal VPA exposure causes a reduction of dopaminergic activity in the prefrontal cortex of male, but not female, mice. This suggests that reduced prefrontal dopaminergic activity is associated with behavioral abnormalities in VPA-treated mice. In the present study, we examined whether the attention deficit/hyperactivity disorder drugs methylphenidate and atomoxetine (which increase dopamine release in the prefrontal cortex, but not striatum, in mice) could alleviate the behavioral abnormalities and changes in dendritic spine morphology induced by prenatal VPA exposure. We found that methylphenidate and atomoxetine increased prefrontal dopamine and noradrenaline release in VPA-treated mice. Acute treatment with methylphenidate or atomoxetine did not alleviate the social interaction deficits or recognition memory impairment in VPA-treated mice, while chronic treatment for 2 weeks did. Methylphenidate or atomoxetine for 2 weeks also improved the prenatal VPA-induced decrease in dendritic spine density in the prefrontal cortex. The effects of these drugs on behaviors and dendritic spine morphology were antagonized by concomitant treatment with the dopamine-D1 receptor antagonist SCH39166 or the dopamine-D2 receptor antagonist raclopride, but not by the α2 -adrenoceptor antagonist idazoxan. These findings suggest that chronic treatment with methylphenidate or atomoxetine improves abnormal behaviors and diminishes the reduction in spine density in VPA-treated mice via a prefrontal dopaminergic system-dependent mechanism. Autism Res 2016, 9: 926-939. © 2015 International Society for Autism Research, Wiley Periodicals, Inc.

Valproic acid (VPA) is a multi-target drug and an inhibitor of histone deacetylase (HDAC). We have previously demonstrated that prenatal exposure to VPA at embryonic day 12.5 (E12.5), but not at E14.5, causes autism-like behavioral abnormalities in male mouse offspring. We have also found that prenatal VPA exposure causes transient histone hyperacetylation in the embryonic brain, followed by decreased neuronal cell numbers in the prefrontal and somatosensory cortices after birth. In the present study, we examined whether prenatal HDAC inhibition affects neuronal maturation in primary mouse cortical neurons. Pregnant mice were injected intraperitoneally with VPA (500 mg/kg) and the more selective HDAC inhibitor trichostatin A (TSA; 500 µg/kg) at E12.5 or E14.5, and primary neuronal cultures were prepared from the cerebral cortices of their embryos. Prenatal exposure to VPA at E12.5, but not at E14.5, decreased total number, total length, and complexity of neuronal dendrites at 14 days in vitro (DIV). The effects of VPA weakened at 21 DIV. Exposure to TSA at E12.5, but not at E14.5, also delayed maturation of cortical neurons. In addition, real-time quantitative PCR revealed that the prenatal exposure to TSA decreased neuroligin-1 (Nlgn1), Shank2, and Shank3 mRNA levels and increased contactin-associated protein-like 2 mRNA level. The delay in neuronal maturation was also observed in Nlgn1-knockdown cells, which were transfected with Nlgn1 siRNA. These findings suggest that prenatal HDAC inhibition causes changes in gene expression of autism-related molecules linked to a delay of neuronal maturation.

These findings indicate that chronic risperidone and aripiprazole treatments improve VPA-induced abnormalities in behaviors and prefrontal dendritic spine density, which may be mediated by repeated elevation of extracellular DA in the prefrontal cortex. Our results also imply that loss of prefrontal dendritic spines may be involved in the abnormal behaviors in the VPA mouse model of ASD.

No abstract

No abstract