Endogenous Retrovirus Expression in Two MOUSE Models of Autism Spectrum Disorders

Background: Mobile retroelements such as Human Endogenous Retrovirus (HERV) have been implicated in many complex diseases with multifactorial etiology and genetic basis, including neurological and neuropsychiatric disorders. In a previous study we demonstrated that specific HERV families show a distinctive expression profile in peripheral blood mononuclear cells (PBMCs) from Autistic Spectrum Disorders (ASD) patients compared to healthy controls, suggesting that HERV expression may contribute to etiopathogenesis of this complex disease.

Objectives: We evaluate the expression profile of different mouse endogenous retrovirus (MERV) families in mouse models of ASD to investigate their potential role in the etiopathogenesis of the disease. Two different ASD mouse models were selected. The first model was fetal exposure to valproic acid (VPA) in outbred mice. VPA is an antiepileptic drug used also for the therapy of bipolar disorders; prenatal VPA exposure in humans is associated with a 7-10x increase in relative risk for ASD, whereas prenatally VPA exposure in mice leads to autism-like anatomical and behavioral abnormalities in the offspring. The second model was the inbred mouse line BTBR T+tf/J (BTBR) that shows behavioral abnormalities analogous to the core ASD symptoms.

 Methods:  We used CD-1 outbred mice prenatally exposed to VPA by subcutaneously injection of a single dose of VPA (500mg/Kg) at gestational day 12.5 and BTBR mice. Behavioural analysis was carried out in developing and adult VPA exposed mice. In both mouse models, blood and brain samples were collected at different postnatal days and relative quantification of several MERV families (ETnI, ETnIIα, ETnIIβ, ETnIIγ, MusD and IAP) were done by Real-time PCR. Standard analysis of variance was applied to assess differences between each group and its control.   

Results: VPA exposed mice showed a delay in acquisition of neonatal motor patterns and reduction of explorative activity at weaning. In blood samples relative expression of MERV families (ETnIIα, ETnIIβ, ETnIIγ, MusD and ETnI) was higher in VPA treated than in controls at all the time of observation (postnatal days 1, 7 and 23). In the brains, MERVs expression differences between VPA treatment and controls were even more evident than in corresponding blood samples. The same enhanced expression profile of MERV was found in the BTBR line: in blood samples and in brain homogenates, relative expression of ETnIIalpha, ETnIIbeta, ETnIIgamma, MusD and IAP, ETnI, and IAP was higher than in controls (C57BL/6j), indicative of higher MERV expression.

Conclusions: Data obtained from blood and brain samples from both BTBR and VPA mouse models support the hypothesis that an increase of the MERVs expression can be associated with the ASD-like behavioural phenotype. These results are in agreement with what we previously described in autistic children. Interestingly the MERV expression profile observed in brain samples from both ASD mouse parallel the one blood samples. The high levels of expression found in brain  pave the way for further studies aimed to clarify whether the different profile of MERV is associated with neuroinflammation and encourage us to verify if drug treatments targeting MERV expression may modify or revert the ASD phenotype.