Morphometric (tessellation) analysis reveals no quantifiable difference in spatial arrangement of map2-immunolabelled neurons in ba32 (cortical layers 5 and 6) in autism

Focal cortical dysplasia and laminar disorganisation are pathological features associated with autism. Recent reports suggest neuronal numbers are reduced in deeper cortical layers (fusiform gyrus and anterior cingulate). Changes affecting neuronal arrangement in these layers would support early migratory disturbances in autism. We analysed the spatial arrangement of MAP2+ neurons in layers 5-6 (L5/6) of the dorsal anterior cingulate cortex (BA32) using a novel tessellation procedure based on the Voronoi method (1). Similar methods have previously been applied to cases with epilepsy showing re-organisational dysplasia (2) and microdysgenesis (3). Postmortem tissue from 8 autism cases (24.0 ± 5.3 years) and 11 controls (28.1 ± 3.9 years) matched for age, gender and hemisphere, were obtained via the Autism Tissue Program (USA) with LREC approval. A 1-in-4 series of sections were immunolabelled to detect MAP2+ neurons (clone HM2, Sigma), and analysed using customised software (Image Pro Plus, Version 5) to determine spatial ‘clustering’ of neurons. The tessellation programme defined the midpoint between a ‘tagged’ neuron and its closest neighbour within a 2-D plane, enabling calculation of coefficient of variation (CV) values of derived polygon areas and mean distances between neighbouring neurons (1-3). L5-6 laminar widths and mean percentage cortical depth values were also compared, but showed no significant differences between cohorts. Total mean CV of polygon areas, CV of polygon class sizes and mean point-to-point distances (between nearest neighbouring neurons) did not differ significantly between groups. At least two of the autism cases in this study showed evidence of focal neuronal crowding and cortical dysplasia elsewhere within the cerebral cortex. However, tessellation analysis failed to reveal any significant differences in spatial arrangements of MAP2+ neurons within deeper cortical layers of BA32. Significant variation in dysplastic features within brain areas and between affected individuals, if present, could be masked when examining cohort/group effects.

1. Duyckaerts, C, Godefroy, G (2000) J Chem Neuroanatomy 20:83-92