An alteration in the balance of excitation-inhibition has been proposed as a common characteristic of the cerebral cortex in autism, which may be due to an alteration in the number and/or function of the excitatory and/or inhibitory cells that form the cortical circuitry. We previously found a decreased number of the parvalbumin (PV)+ interneuron known as Chandelier (Ch) cell in the prefrontal cortex in autism. This decrease could result from a decreased number of Ch cells, but also from decreased PV protein expression by Ch cells. To further determine if Ch cell number is altered in autism, we quantified the number of Ch cells following a different approach and different patient cohort than in our previous studies. We quantified the number of Ch cell cartridges—rather than Ch cell somata—that expressed GAT1—rather than PV. Specifically, we quantified GAT1+ cartridges in prefrontal areas BA9, BA46, and BA47 of 11 cases with autism and 11 control cases. We found that the density of GAT1+ cartridges was decreased in autism in all areas and layers. Whether this alteration is cause or effect remains unclear but could result from alterations that take place during cortical prenatal and/or postnatal development.

Fragile X-associated tremor/ataxia syndrome (FXTAS) is a neurodegenerative disorder that affects older premutation carriers (55–200 CGG repeats) of the fragile X gene. Despite the high prevalence of the FXTAS disorder, neuropathology studies of individuals affected by FXTAS are limited. We performed hematoxylin and eosin (H&E) staining in the hippocampus of 26 FXTAS cases and analyzed the tissue microscopically. The major neuropathological characteristics were white matter disease, intranuclear inclusions in neurons and astrocytes, and neuron loss. Astrocytes contained more and larger inclusions than neurons. There was a negative correlation between age of death and CGG repeat length in cases over the age of 60. The number of astroglial inclusions (CA3 and dentate gyrus) and the number of CA3 neuronal inclusions increased with elevated CGG repeat length. In the two cases with a CGG repeat size less than 65, FXTAS intranuclear inclusions were not present in the hippocampus, while in the two cases with less than 70 (65–70) CGG repeat expansion, neurons and astrocytes with inclusions were occasionally identified in the CA1 sub-region. These findings add hippocampus neuropathology to the previously reported changes in other areas of the brain in FXTAS patients, with implications for understanding FXTAS pathogenesis.