A Deficit of Long-Range Connectivity Due to Corpus Callosum Hypoplasia Is Present in Idiopathic and Syndromic (dup15) Autism

Background: A reduced size of the corpus callosum (CC) is a common feature reported in neuroimaging studies in individuals diagnosed with autism (Minshew 2007, Frazier 2009) suggesting a deficit of long distance connectivity. Postmortem studies of the CC in idiopathic autism confirmed a reduced midsaggital area and explained neuroimaging findings by demonstrating a deficit of the total number of CC axons by 48.4% and axonal numerical density by 37% (Wegiel et al, submitted). We hypothesize that CC hypoplasia is a common marker of congenital abnormalities in autism regardless of disorder etiology.

Objectives: The aim of this postmortem study was to establish the type and topography of structural CC anomalies in individuals with dup(15), to compare patterns of CC pathology in dup(15)/autism and idiopathic autism, and to determine the contribution of this pathology to the deficit of long-distance connectivity and behavioral anomalies in these two groups.

Methods: Based on the results of clinical, genetic and neuropathological evaluations, eight brains of individuals with dup(15) were included in this study. Our preliminary evaluation had confirmed the diagnosis of dup(15) and autism, and revealed a high prevalence of epilepsy, microcephaly, focal defects of neuronal migration and dysplasia (Wegiel et al 2012). The brain hemispheres of eight individuals 5 to 39 years of age diagnosed with dup(15)/autism and 8 age-matched control subjects were fixed with formalin, dehydrated, embedded in polyethylene glycol, and cut into equidistant serial 50-µm thick sections. They were used to estimate the midsaggital area of the corpus callosum and the total number and numerical density of axons in five CC segments determined using Hofer and Frahm (2006) CC segmentation, which reflects the cortical distribution of neurons involved in CC formation and function.

Results: There was microcephaly with 20% less brain weight in dup(15)/autism than in control group (1,094g and 1354 g, respectively). Although the length of the CC was almost identical in these two groups (58 mm and 60 mm, respectively), the midsaggital area was 40% less in dup/15/autism than in control individuals (253 mm² and 419 mm², respectively). The total number of CC axons was 73% less in dup15/autism than in the control group (22 and 83 million, respectively) whereas the numerical density of axons (N/mm²) was 52% less in the affected subjects than in control individuals (89,000 and 188,000, respectively). The significant deficit of the number of axons in all five CC segments corresponds to a severe deficiency of long distance connectivity of the prefrontal (segment I), premotor (II), motor (III), primary sensory cortex (IV), and the parietal, temporal and occipital cortex (V) in subjects with dup(15)/autism.

Conclusions: Finding similar patterns of developmental anomalies in dup(15)/autism and idiopathic autism indicate that reduced sizes of the CC are consistent markers of congenital brain defects in both autism with genetic and idiopathic etiologies. The results suggest that dup(15) has a direct contribution to both CC hypoplasia and the autistic phenotype.