Social Gestures Are Differentially Related to Brain GABA Levels between Children with Autism and Typically Developing Children

Background: Autism Spectrum Disorder (ASD) is characterized by impairments in social cognition, communication, and repetitive behaviors. Motor abnormalities are common in ASD and have been linked to social and communicative features of the disorder. Praxis - the performance of complex gestures with a communicative or functional purpose, has been shown to be impaired in ASD and has a large impact on quality of life. The inhibitory neurotransmitter GABA plays an important role in regulating motor behaviors and is implicated in ASD pathophysiology, but the link between GABA and autism-related dyspraxia is unclear.

Objectives:Here, we assessed praxis and measured GABA levels in the sensorimotor cortex of typically developing children and children with ASD using edited Magnetic Resonance Spectroscopy. We hypothesize that children with ASD show worse praxis, and that better praxis performance is correlated with higher sensorimotor GABA levels.

Methods: Data were acquired in 24 children with ASD (10.52 ± 1.25 years, 6F) and 26 TDC (10.11 ± 1.38 years; 8F). Subject and parental consent were obtained under local IRB approval. Diagnosis was confirmed using the ADOS-2. Praxis: A pediatric version of the Florida Apraxia Battery was used to examine performance of skilled gestures, known to robustly discriminate children with ASD and TDC. Total percent correct and total error scores were used. MRI: GABA-edited MR spectra were acquired from (3 cm)³ voxels over the right primary sensorimotor cortex using MEGA-PRESS on a Philips 3T ‘Achieva’ (TE/TR 68/2000 ms, 320 transients; 10 min) and analyzed in Gannet.

Results: Mean praxis percentage correct (ASD; 52.1 ± 15.21, TDC; 75.93 ± 10.94, p < 0.001) and errors (ASD; 42.3 ± 15.95, TDC 20.57 ± 9.87, p < 0.001) differed significantly. GABA values were significantly lower in ASD, as previously shown. TDC showed a negative correlation with praxis percentage correct (-0.62, p = < 0.01), whereas there was trend towards a positive correlation in ASD (0.21, p = 0.12; Figure 1A). TDC showed a positive correlation between GABA and praxis errors (R – 0.6, p < 0.001; Figure 1B). Children with ASD showed a negative trend (R = -0.25). The correlations were significantly different between cohorts (Fisher r-to-z, p < 0.01). We explored whether an inverted u-shape (quadratic fit) fit the data across all participants, but significance did not reach alpha of 0.05 (R² = 0.078, p =0.06; Figure 1C). There were no significant differences in data quality.

Conclusions: GABA is differentially related to praxis performance in children with ASD and TDC. It is not quite clear how to interpret this finding, but the correlations are robust in TDC. Additional resting-state-fMRI, or left-hemisphere GABA levels, might elucidate this relationship. Within TDC, increased GABA levels may reduce gesture performance by suppressing SMA/PMC function to properly plan and execute movements, whereas in ASD, pathologically-reduced GABA may hinder efficient encoding of motor commands and ultimately impair the performance of complex gestures with a communicative or functional purpose, contributing to the autistic phenotype. Understanding the GABA system in ASD is important for developing future novel targets for patient-specific treatment.