Increased Iba-1 Positive Microglial Cell Density and Somal Volume in the Autistic Dorsolateral Prefrontal Cortex

Background:

Autism is marked by early brain overgrowth; one brain region in which overgrowth is strongly pronounced is dorsolateral prefrontal cortex (DLPFC). Several studies have reported changes in gene and protein expression levels consistent with pro-inflammatory changes in the autistic brain. Qualitative observations and a non-significant trend increase in HLA-DR staining have suggested the possibility of microglial activation in the frontal cortex of adults and older children with autism.

Objectives:

Determine whether there are significant alterations in neuronal organization, local microglial-neuronal clustering, microglial density, and microglial volume in the DLPFC in autism using an antibody to ionized calcium binding adaptor molecule-1 (Iba-1), which images both activated and resting microglia. Assess whether alterations are present at the youngest ages, and whether there is any interaction with the presence of seizure.

Methods:

Tissue from the DLPFC of n=16 autistic and n=10 control hemispheres was sectioned at a thickness of 50 microns.  Histology was performed in 8 gyri per case using the Iba-1 antibody at 1:1000 concentration with a hematoxylin-eosin counterstain.  Density counts were performed for both gray and white matter via a 1.4 n.a. 100x objective using a 3-d optical disector with a 2 micron apical guard zone at sufficient sampling density to achieve a Schaeffer CE <.10 for each case.  Volume was projected for all counted microglia from 2-d isotropic nucleator.  Coordinate locations of microglia and neurons were collected at a high sulcal wall location over an area no less than 625 microns in width and extending from pial surface to white matter.  A density-independent univariate spatial pattern analysis was performed on these coordinates, along with a bivariate spatial pattern analysis over a 35 micron range and a density-independent assessment of the average distance from microglia to their nearest neuronal neighbors.

Results:

Significant increases in microglial density were observed in autism in both gray (25%; p<.001) and white (17%; p<.01) matter. Significant increases in microglial volume were observed in autism in both gray (30%, p=.04) and white (43%, p=.01) matter. A significant increase in neuronal clustering was present (p=.05), along with an increase local microglial-neuronal clustering (p=.01) and decline in the microglial-neuronal nearest-neighbor distance (p=.02). In autistic subjects under 6 years of age (n=3), percent changes in microglial density and volume were remarkably similar to those observed in subjects over 6 years of age. The majority of microglial measures demonstrated greater alterations in the seizure-free autistic group than the seizure group. No consistent interaction with postmortem interval was observed.

Conclusions:

Increases in microglial density, microglial somal size, neuronal clustering, and local microglial-neuronal clustering are present in the DLPFC in autism and do not appear to primarily reflect effects of seizure or postmortem interval. The observed changes are reminiscent of inflammatory reactions in other disorders, but might also reflect aberrant microglial genesis or abnormal, non-inflammatory microglial activity in the developing autistic brain. The observation of microglial alterations in young autistic subjects that are similar to those in older subjects suggests that microglial abnormality may be one feature underlying early aberrant brain overgrowth.