Increased Neuron Number and Head Size in Autism

Abstract

N THIS ISSUE OF JAMA, THE REPORT BY COURCHESNE AND colleagues 1 documents an increase in neuron number intheprefrontalcortex(PFC)inmalechildrenandadolescents with autism. Those findings build on Leo Kanner’s original observations 2 in 1943 and 2 decades of recent research investigating macrocephaly in autism. Macrocephaly occurs in 20% of individuals with autism on average and is usually due to megalencephaly—abnormal enlargement of the brain during childhood. 3 The enlargementisrarelypresentatbirth;itdevelopsduringearlychildhood when head growth accelerates during the first 18 months of life. 4 Mean total brain, lobar, white matter, and gray matter volumes, including volume of the cortex, are significantlyincreasedby2to3yearsofageinchildrenwith autism when compared with typically developing and also nonautistic developmentally delayed individuals. 5-7 The excessive head growth and brain growth occur prior to most clinical manifestations of the disorder, raising the possibility that the mechanisms that cause excessive growth also play a role in the primary developmental neuropathology of autism. The purpose of the study by Courchesne et al 1 was to investigate the neural underpinnings of brain overgrowth in autismat the cellular level. In this postmortemtissue study, the investigators focused on cortical gray matter, specifically the PFC, which is one of the brain regions highly implicatedinthedisorder,andsoughttoansweracriticalquestioninautismresearch: is brain overgrowthassociated with increased size and number of neurons, glial cells, or both? The investigators found 79% more neurons in the dorsolateral PFC (DL-PFC) and 29% more neurons in the mesial PFC (M-PFC) in 7 brains from boys with autism compared with 6 brains of typically developing children (controls). There were no apparent group differences in neuron size or in glial cell counts. Prefrontal neuron number and brain weight were significantly correlated in the control group but not in the autism group. The authors speculated that the latter finding may indicate that the brains of some individuals with autism are undersized relative to their prefrontal cortical neuron count. Equally possible is an increaseofprefrontalcorticalneuronnumberanddensityrelativetotypicalbrainweight.Thebrainsofautismcases3and 4 in the study by Courchesne et al 1 are specific examples of such an alternative explanation, having increased numbers of neurons in the PFC (2.21 and 2.18 billion; eTable 5 in the article 1 ) and weights within the range of control brain weights. Postmortem brain tissue studies in autism research, and in general, are difficult to conduct due to limited availability of brain samples, non-representativeness of the samples, variable causes of death, potentially long postmortem intervals, and differences in brain extraction, sectioning and tissueprocessingprotocolsandcellcountingmethods. 8 Most if not all case-control tissue sample sizes are small, and representative population-based postmortem reference data, similar to those currently available from large normative samples of in vivo pediatric brain imaging, 9 do not exist.