Reduced Auditory Cortical GABA Concentration in ASD and First Degree Relatives

Background:  

One emerging perspective on the pathophysiology of Autism Spectrum Disorders (ASD) is the excitation/inhibition imbalance (EI) theory, which proposes that relatively high ratios of excitatory to inhibitory neuronal processes could explain some portion of the ASD phenotype. Evidence for inhibitory deficits converges from a variety of methods and has been of interest for some time, and is supported by a significant body of evidence, including changes in receptor expression. Messenger RNA levels of glutamate decarboxylase (GAD), the enzyme that converts glutamate to GABA, is reduced in ASD, suggestive of corresponding changes in GABA. To date, there has only been a single study of GABA in ASD using MRS methods, which reported that GABA concentration was lower in the frontal lobe in ASD.

Objectives:  

Our current study was aimed at the auditory cortex, where we have previously reported electrophysiological abnormalities in gamma-band oscillations in ASD suggestive of changes in EI. We were also interested in whether results would extend to first-degree relatives of persons with ASD, and predicting lower GABA levels in relatives because we have seen changes in EI balance our gamma-band studies of relatives.

Methods:  

GABA levels were determined from the left superior temporal gyrus for 9 ASD participants, 12 siblings of persons with ASD (SIB), 11 parents of persons with ASD (pASD) and 11 healthy controls (HC). A MEGAPRESS spectral editing sequence was used, optimized for GABA detection, with edit-on and edit-off frames interleaved, for a total of 512 acquisitions (256 edit-on and 256 edit-off frames), on a GE 3T HDx scanner with TR/TE= 2500ms/70ms.  Edit-on and edit-off frames for each acquisition were separated, reconstructed, and fitted using SAGE (GE Healthcare). The processed edit-off spectra were subtracted from the edit-on spectra to produce the GABA spectra.  The GABA peak area was divided by the area of the creatine peak in the edit-off spectra to produce the GABA/Cr ratio. 

Results:  

The GABA/Cr ratio was entered into a one-way ANCOVA with group as the independent variable. Age was the single covariate chosen because it was significantly correlated with the dependent measure. The main effect of group was significant, F(3,38) = 3.51, p < .02. Post-hoc LSD analyses revealed the following significant differences in GABA/Cr ratios: HC > ASD (p = .02), HC > pASD (p = .04), and HC > SIB (p = .038).

Conclusions:  

The significantly reduced auditory cortex GABA in the ASD group provides support for the hypothesis of GABAergic inhibitory deficits in ASD, consistent with one earlier report of GABA reduction in the frontal lobe in ASD.  The fact that reductions were also present in the two groups of 1^st degree relatives suggests that the GABA reduction may be heritable as well, and could be used as an endophenotype in future ASD studies. Although these data provide support for the inhibitory side of EI theory, future studies should include both MEGAPRESS and short-echo PRESS measurement of glutamate from the same voxels in participants.