Dysgenesis of the Corpus Callosum As Evidence of Developmental Defects of Long Distance Connectivity in Autism

Background: The corpus callosum (CC) is the largest white matter tract containing 138 to 195 million axons (Riise 2006, Tomasch 1954) that connect homologous brain cortical areas. Neuroimaging studies reveal a reduced size of the CC in individuals diagnosed with autism (Minshew 2007, Frazier 2009) whereas postmortem studies of control subjects reveal a correlation between the area of CC and the number of axons (Riise and Pakkenberg 2006). Despite strong neuroimaging evidence of CC pathology, the microscopic characteristics of axonal CC abnormalities in autism are lacking. 

Objectives: The aim of this postmortem study of the brain of autistic subjects is (a) to characterize the type and topography of structural CC defects and (b) expand the concept of CC pathology by detection of markers of focal CC midline axonal misguidance.  

Methods: The brain hemispheres used in this study were preserved for several parallel studies designated to detect the global pattern of developmental abnormalities in autism. Brain hemispheres of 10 autistic individuals from 4 to 56 years old and 10 age matched control subjects were fixed with formalin, dehydrated, embedded in celloidin, and cut into serial 200-µm thick sections that were stained with cresyl violet. Every 6^th section was used to establish the midsaggital area of the corpus callosum. Postmortem MRI and serial coronal sections were used to characterize CC developmental abnormalities. Segmentation of the CC applied in this study was based on diffusion tensor imaging and fiber tractography distinguishing five CC regions reflecting the position of cortical neurons contributing to CC connectivity (Hofer 2006).  

Results: This postmortem study revealed (a) more CC shape differences, (b) significant reductions of the midline CC area (p<0.05) associated with significant area deficits in regions II, IV and V (p<0.05), and (c) partial agenesis of the corpus callosum in autistic subjects. A consistent feature in three cases was the lack of axonal connections in region IV (primary sensory cortical connections), lack in one case and a severe (50% and 66%) deficit in the other two cases of axonal connections in region III (primary motor cortex connections). In addition in an 11 year old female, a 60% deficit was detected in region I (prefrontal cortex projections) and a 42% deficit in region V (connections of parietal, temporal and occipital cortex). CC dysplasia detected in autistic subjects, a several millimeter-wide gap between truncated arms of the left and right corpus callosum, appears to be a marker of axonal misguidance in the CC midline. Studies of animal models and human fetal brain suggest that the axonal misguidance detected in examined autistic subjects reflects a developmental failure of guidepost cells including glial wedge, indusium griseum, midline zipper glia, and subcallosal sling cells (Richards 2004, Sanchez-Camacho 2011).  

Conclusions: Regional reduction of CC area and midline CC dysgenesis reflect deficits of axonal interhemispheric connections and defects of long distance connectivity in autism. Our study demonstrates the presence of cortical neurons interhemispheric connectivity deficit in the majority of postmortem examined autistic subjects and focal CC midline axonal misguidance in 30% of cases.