The cognitive phenotype of autism has been correlated with an altered balance of excitation to inhibition in the cerebral cortex, which could result from a change in the number, function, or morphology of GABA-expressing interneurons. The number of GABAergic interneuron subtypes has not been quantified in the autistic cerebral cortex. We classified interneurons into 3 subpopulations based on expression of the calcium-binding proteins parvalbumin, calbindin, or calretinin. We quantified the number of each interneuron subtype in postmortem neocortical tissue from 11 autistic cases and 10 control cases. Prefrontal Brodmann Areas (BA) BA46, BA47, and BA9 in autism and age-matched controls were analyzed by blinded researchers. We show that the number of parvalbumin+ interneurons in these 3 cortical areas-BA46, BA47, and BA9-is significantly reduced in autism compared with controls. The number of calbindin+ and calretinin+ interneurons did not differ in the cortical areas examined. Parvalbumin+ interneurons are fast-spiking cells that synchronize the activity of pyramidal cells through perisomatic and axo-axonic inhibition. The reduced number of parvalbumin+ interneurons could disrupt the balance of excitation/inhibition and alter gamma wave oscillations in the cerebral cortex of autistic subjects. These data will allow development of novel treatments specifically targeting parvalbumin interneurons.

The observed behavioral deficits, especially regarding social behaviors, strengthens the face validity of the gabrb3 gene deficient mouse as being a model of autism spectrum disorder.

We report quantitative label-free imaging with phase and polarization (QLIPP) for simultaneous measurement of density, anisotropy, and orientation in unlabeled live cells and tissue slices. We combine QLIPP with deep neural networks to predict fluorescence images of diverse cell and tissue structures. QLIPP images reveal anatomical regions and axon tract orientation in prenatal human brain tissue sections that are not visible using brightfield imaging. We report a variant of UNet architecture, multi-channel 2.5D U-Net, for computationally efficient prediction of fluorescence images in three dimensions and over large fields of view. Further, we develop data normalization methods for accurate prediction of myelin distribution over large brain regions. We show that experimental defects in labeling the human tissue can be rescued with quantitative label-free imaging and neural network model. We anticipate that the proposed method will enable new studies of architectural order at spatial scales ranging from organelles to tissue.

The title and the penultimate paragraph of the Discussion section incorrectly referred to the "medial prefrontal cortex." The word "medial" should not have appeared. This has been corrected in the article online.

Autism spectrum disorder (ASD) is diagnosed based on three core features: impaired social interactions, deficits in communication and repetitive or restricted behavioral patterns. Against this backdrop, abnormal sensory processing receives little attention despite its prevalence and the impact it exerts on the core diagnostic features. Understanding the source of these sensory abnormalities is paramount to developing intervention strategies aimed at maximizing the coping ability of those with ASD. Consequently, we chose to examine whether sensory abnormalities were present in mice heterozygous for the Gabrb3 gene, a gene strongly associated with ASD. Mice were assessed for tactile and heat sensitivity, sensorimotor competence (accelerating rotarod task) and sensorimotor gating by prepulse inhibition of the acoustic startle reflex (PPI). All heterozygotes exhibited an increase in seizure susceptibility and similar reductions in Gabrb3 expression in the dorsal root ganglia, spinal cord, whole brain and amygdala. Interestingly, significant differences were noted between heterozygous variants in regards to tactile sensitivity, heat sensitivity, sensorimotor competence and PPI along with differences in Gabrb3 expression in the reticular thalamic nucleus and the bed nucleus of stria terminalis. These differences were influenced by the heterozygotes’ gender and whether the Gabrb3 gene was of paternal or maternal origin. These results are not adequately explained by simple haploinsufficiency of Gabrb3, therefore, additional mechanisms are likely to be involved. In addition, this is the first report of the occurrence of tactile and heat hypersensitivity in an ASD mouse model, two features often associated with ASD.