Abstract

Autism spectrum disorders (ASD) comprise a genetically heterogeneous group of neurodevelopmental disabilities characterized by repetitive behaviors as well as deficits in communication and social/emotional interactions and behaviors. Defects in GABA transmission have been hypothesized to underlie the symptoms of ASD (Ben-Ari, Khalilov, Kahle, & Cherubini, 2012). Engrailed-2 (En2) is a homeodomain transcription factor involved in regionalization and patterning of the midbrain and hindbrain regions (Joyner, 1996; Gherbassi and Simon, 2006). En2 is expressed in the developing and adult mouse midbrain/hindbrain region (Joyner, 1996), as well as the adult mouse forebrain (Tripathi et al., 2009). Genome wide association studies have shown that two intronic single-nucleotide polymorphisms in the human En2 gene are significantly associated with ASD (Benayed et al., 2009). En2-/- mice lack the homeodomain of En2 (En2hd/hd mice; Joyner et al., 1991, referred to as En2-/-) and display neuropathological and behavioral changes relevant to ASD. We first investigated the distribution of different interneuron subtypes in the cerebral cortex and hippocampus of En2-/- mice. Results showed a significant loss of parvalbumin, somatostatin and neuropetide Y positive interneurons, in both regions of En2-/- brains. Brain-derived neurotrophic factor (BDNF) is a crucial factor for the postnatal development of forebrain GABAergic neurons, namely PV interneurons. We therefore investigated whether altered BDNF expression may be related to the altered development of GABAergic interneurons, described in En2-/- mice. Considering the important role of BDNF in forebrain GABAergic interneuron development, we decided to determine whether interneuron loss in En2-/- forebrain might be related to altered expression of BDNF and its signaling receptors. We evaluated the expression of different BDNF mRNA and protein isoforms in various brain areas from wild-type (WT) and En2-/- mice. Quantitative RT-PCR indicated an alteration of various splicing variants of BDNF mRNA in the cerebral cortex but not hippocampus and cerebellum of adult En2-/- mice, as compared to WT controls. Immunoblot analyses instead revealed increased levels of mature BDNF and reduced levels of truncated- and pro-BDNF isoforms in the hippocampus of En2-/- vs. WT mice. These data indicate a role of BDNF in the anatomical phenotype of En2-/- mice, suggesting that BDNF might be involved in the developmental defects of the GABAergic system in this mouse model of ASD. We next investigated the effect of BDNF on GABAergic and glutamatergic neuron connectivity using primary cultures of hippocampal neurons from WT and En2-/- mice. Results showed that BDNF is able to increase the number of GABAergic and glutamatergic synaptic contacts in En2-/- hippocampal neurons. Taken together, our results suggest a beneficial role of BDNF in rescuing synaptic defects in En2-/- mice.