Skip to main content
Premium Trial:

Request an Annual Quote

Age-Related Gene Expression Differences in Autism Detected in Post-Mortem Brain Samples

By Andrea Anderson

NEW YORK (GenomeWeb News) – A study appearing online last night in PLoS Genetics has found gene expression differences in brain samples from children with autism spectrum disorder compared to adults with the condition, hinting that different pathological processes may be at work in the autistic brain depending on age.

A California research team used arrays to assess almost three-dozen post mortem brain samples from individuals with or without autism, gauging both gene expression and copy number alterations in a region called the prefrontal cortex. Along with comparisons between unaffected and affected individuals, the investigators looked at how expression patterns in samples from toddlers and children with ASD compared with those in samples from adolescents and adults with ASD.

Though some gene expression patterns in the prefrontal cortex turned up regardless of age in the ASD group, researchers also found age-dependent gene expression alterations in autism. For example, samples from children with ASD tended to show atypical expression of genes contributing to neuron number — consistent with some of the unusual early brain growth patterns previously described in ASD — whereas adult autism cases were marked by unusual expression profiles involving genes from signaling, immune, and repair pathways.

"We're showing that the adult condition — where you have neuron loss and you have immunological activation — isn't the way it began. It's an outcome," co-corresponding author Eric Courchesne, director of the National Institutes of Health-University of California at San Diego's Autism Center of Excellence, told GenomeWeb Daily News.

For more than a decade, studies have been finding distinct early brain development patterns in autism, including excess neurons in certain parts of the brain. This overgrowth seems to stop sometime during childhood, Courchesne noted. And by adolescence or adulthood, the enlarged brain regions seen in childhood generally seem to disappear through processes such as neuron loss and/or thinning in some cortical areas.

That has made it tricky to figure out the molecular roots of the early brain overgrowth seen by neuroimaging experiments or postmortem analyses, authors of the new study explained, though previous genetic studies have garnered clues about some of the age-independent processes contributing to autism.

"Up until now, there has been no information on the developmental, molecular pathology of autism — that is, what's going wrong early in the brain that's responsible for excess and abnormal brain growth," Courchesne explained.

He and his colleagues used Illumina microarrays to assess expression profiles across the genome in post-mortem prefrontal cortex samples, comparing the patterns in unaffected and affected individuals and in younger and older individuals with ASD.

Because they were dealing with RNA from post-mortem brain samples, which are often not collected immediately after death, the researchers sent samples to Illumina for processing with its DASL assay to help get as much signal as possible out of the microarray experiments.

The team focused on the prefrontal cortex in part because it is one of the brain regions where early neuronal overgrowth seems to be most pronounced. The brain region is also believed to be an important mediator of some of the same processes affected in autism, including communication and social interactions, emotion, language, and complex behavior regulation.

Of the 33 samples included in the study, nine came from boys with ASD who were between the ages of two and 14 years old and seven came from unaffected boys in the same age group. Six represented males with ASD in the 15 to 56-year-old age range and 11 more came from adult male controls.

When they analyzed their expression data, researchers saw that the expression profiles for some genes were altered regardless of age in autism, including some found in cell cycle checkpoint or cytoskeletal pathways. Likewise, ASD-related copy number changes detected by genotyping 55 post-mortem prefrontal cortex samples often affected genes implicated in autism in the past.

On the other hand, the researchers found hundreds of genes with distinct expression profiles in prefrontal cortex samples from children with ASD compared to those from adults with the condition.

Children with ASD often showed differential expression of genes contributing to neuron formation, differentiation, and cortical patterning, they reported, along with genes that may influence how extra neurons are weeded out as connections between neurons are formed and shored up in the developing brain.

"The main pattern that jumped out involved early developmental molecular networks," Courchesne said. "In early development, the first steps involve the genesis and regulation of neuron numbers — how many neurons and how many do you retain?"

For adults with ASD, though, researchers found that the expression changes that stood out most in the prefrontal cortex samples involved signaling and immune genes as well as genes from neuron removal, remodeling, and repair-related pathways.

"It's a plausible explanation that, secondarily, the neurobiology of a brain that has an excess number of synapses, neurons, and connections has to go through a prolonged period of remodeling and restructuring," Courchesne said. Even so, he emphasized that that is not something that has been directly demonstrated so far.

More research will be needed to explore that possibility and to determine when and how the shift in expression patterns occurs in the prefrontal cortex during autism, he added.

Courchesne and his colleagues are also interested in exploring the gene expression patterns that characterize childhood and adulthood in other brain regions and plan to do further studies of post-mortem samples to look at some of the other brain sites where brain overgrowth in childhood has been reported.

From their findings in the prefrontal cortex, combined with information from past imaging studies, the researchers expect that they might see differences in the nature and extent of the gene expression abnormalities in autism depending on the brain area examined.

While the studies described in the current paper hinged on array-based expression analyses, the team is now turning to sequencing-based approaches for profiling post-mortem brain tissue.

In the longer term, the researchers hope to figure out whether the gene expression alterations detected in the brains of children with autism are reflected by any characteristic blood-based gene expression profiles. If so, Courchesne explained, it might eventually be possible to use such profiles as blood-based markers to identify autism cases very early on in childhood.

"Blood cells are a different tissue type, so there will be [expression] differences," he said. "Nonetheless, there's an interesting possibility that if, at very young ages, blood was examined for gene expression profiles, some signals might be found that are in common with the brain tissue signals."