The Contribution of Reduced TrkB Signaling to Autistic-like Behaviour in the Valproic Acid-Induced Mouse Model

Background:  The molecular mechanisms underlying autistic-like behaviour remain to be elucidated. Mutations in autism-linked genes adversely affect molecules, such as brain-derived neurotrophic factor receptor TrkB and Akt/mTOR pathway components, which regulate dendritic spine formation, function and plasticity. In line with these findings, we previously demonstrated reduced TrkB/Akt/mTOR protein and signaling in human idiopathic autism and in the valproic acid (VPA)-induced rodent model of autism. These results support the hypothesis that defective TrkB/Akt/mTOR pathway signaling is a molecular substrate of autistic behaviour and identifies this pathway as a potential therapeutic target for autism. Promisingly, decreased TrkB signaling via Akt has been rescued by administration of the TrkB partial agonist LM22A-4 in an animal model of Rett syndrome, a monogenic disorder associated with high rates of autism.

Objectives:  To better establish the contribution of decreased TrkB/Akt/mTOR pathway signaling to autistic behaviour, we aimed to examine whether systemic treatment with the partial TrkB agonist LM22A-4 would restore TrkB/Akt/mTOR signaling and ameliorate autistic-like behaviour in the offspring of VPA-injected mothers.

Methods:  Pregnant CD1 mice received a single intraperitoneal (i.p.) injection of 500 mg/kg VPA on gestational day 12.5, while controls were injected with only saline. Pups were weaned on postnatal day (PD) 21 and received an i.p. injection of either saline or LM22A-4 (0.05mg/g) once daily from PDs 21-35. Sociability and repetitive/stereotypic behaviour were evaluated on PDs 29-34 using the three-chambered social approach task and marble-burying test, respectively. Litters were killed and brain tissue harvested on PD 35. Protein expression and phosphorylation levels of the serine/threonine protein kinase Akt, a key downstream effector of TrkB, were measured by Western blotting in the temporal/parietal neocortices of saline- and LM22A-4-treated VPA and control mice.  

Results:  Behavioral results were dependent on sex, with female VPA-exposed mice displaying a lack of sociability and male VPA-exposed mice displaying increased repetitive behaviour compared to controls. Contrary to VPA-exposed saline-treated females, LM22A-4-treated VPA-exposed female mice displayed normal sociability. Also, VPA-exposed male mice treated with LM22A-4 buried fewer marbles than saline-treated VPA-exposed males, consistent with a decrease in repetitive behaviour. Lastly, both LM22A-4-treated VPA-exposed females and males had higher levels of phosphorylated Akt than saline-treated VPA-exposed mice, similar to untreated controls. 

Conclusions:  Our results support the hypothesis that reduced TrkB/Akt/mTOR signaling contributes to autistic behaviour and that this pathway might have a therapeutic role in treating idiopathic autism. Indeed, promisingly, we show that systemic treatment with the partial TrkB agonist LM22A-4 appears to restore the reduction in TrkB signaling and improve social and repetitive behaviours in the VPA mouse model of autism.