News The latest developments in autism research.
Profiles Portraits of scientists who are making a mark on autism research.
Toolbox Emerging tools and techniques that may advance autism research.
Spotted A roundup of autism papers and media mentions you may have missed.
Opinion Conversations on the science of autism research.
Viewpoint Expert opinions on trends and controversies in autism research.
Columnists Dispatches from experts on various facets of autism.
Crosstalk Debates and conversations about timely topics in autism.
Reviews Exploring the intersection of autism and the arts.
Q&A Conversations with experts about noteworthy topics in autism.
Deep Dive In-depth analysis of important topics in autism.
Special Reports Curated collections of articles on special topics in autism.
Webinars Presentations by leading experts on their latest research.
Opinion / Cross Talk

Autism lesion or artifact? Experts discuss new find

by  /  30 April 2014

In early April, we reported on a study that found distinct areas of seemingly immature neurons in the brains of children with autism. The study analyzed gene expression in postmortem brain tissue and identified patches of disorganized cells in the brain’s cortex, or outer layer.

The results suggest that autism arises in utero while the cortex is still developing. The study garnered widespread coverage in the popular press and spurred debate in the research community about its methods, the significance of the findings and how best to confirm them.

We asked experts who study brain pathology and development for their perspectives on the study.

What do you think? Share your reactions and follow-up questions in the comments section below.

The Experts:

Flora Vaccarino

Professor, Yale University

Robert Hevner

Professor, Seattle Children's Research Institute - Seattle Children's Hospital

Margaret Esiri

Fellow of the Royal College of Pathologists, Professor of Neuropathology, Oxford University

Findings leave process unclear

Harris Professor, Child Study Center, Yale University

Higher-order hints: “This new study adds more meat to the emerging evidence that crucial aspects of embryonic brain development are disrupted in autism. Their suggestion is that the morphogenesis of cortical layers, a process that occurs in the first trimester of gestation in humans, is fundamentally disorganized.

“Other reports (that is, from the labs of Manuel Casanova and Daniel Geschwind) have suggested that there are area-specific cortical disruptions of gene expression in the disorder, again pointing to subtle alterations in fundamental processes governing the way the brain is initially built. In this new paper, gene expression was disrupted in the patches in a coordinated fashion (that is, for more than one gene and in more than one layer), suggesting that some kind of higher-order process is disrupted, rather than its being a problem with isolated gene expression.”

Too soon: “I consider this report preliminary because I would have expected to see parallel evidence in the tissue analysis of disrupted layers in the cortex, as commonly is seen in disorders of neuron migration. Cortical layers are each formed by specific neuron types, each with a typical shape, size and connectivity, specified by programs of interacting genes expressed in their progenitor cells.

“Assuming that the neurons are abnormal, and that this is not due to some artifact altering RNA levels in these patches, the paper leaves unclear the process that is disrupted in these brains: Are the neurons mis-specified to a non-cortical, non-layer neuron type? Or is their migration altered, such that the cortical neurons are there, but their position is altered?”

Needs confirmation

Professor of Neurological Surgery, Seattle Children’s Research Institute, University of Washington

Compelling statistic: “The ‘patches’ of disorganization in this study were found in 10 out of 11 brains with autism, but only 1 out of 11 control brains, suggesting that the patches are, if not specific, at least highly enriched in most autism brains — which is surprising given the etiological, genetic and phenotypic heterogeneity of the disorder. Indeed, despite the small, explorative nature of the study, this statistical difference is its most compelling aspect, because the significance of such molecular patches is otherwise completely unknown.”

Important caveats: “First, it is notable that neurons migrated to the correct layer and grew to the correct size and shape, but presumably failed to then express the correct layer-specific molecular profile. Second, there is as yet no evidence that the observed molecular defects had any impact on neural connectivity or physiology, or that they relate in any way to autism phenotypes. Third, no mechanism that would lead to the formation of patches during development is apparent.”

Handle with care: “Finally, the possibility that the patches are mere artifacts arising focally in the tissue after death is far from excluded. In fact, it is well known that RNA degrades rapidly in postmortem tissue. The authors tried to address this possibility by also studying the expression of genes specific for interneurons and glia, which were seen to be mostly intact in patches (although there were exceptions). However, because projection neurons exhibit selective vulnerability to many pathological insults, they may also be more vulnerable to artifact, given their large size and unique membrane properties. Even gentle handling can cause anomalies in the cellular architecture of the cortex, and might conceivably disrupt cells in subtle ways that accelerate molecular degradation.”

Random differences? “Why would such artifacts be more prevalent in the brains of individuals with autism than in those of controls? Here, it is important to remember that the number of cases was small and differences could arise randomly. In addition, it is likely that the autism brains were simply subjected to more handling during the postmortem examination and sampling.”

“Overall, this work demonstrates a new type of possible lesion that may have tremendous importance in autism, but should be viewed with caution until independent confirmation is obtained and the possibility of experimental artifacts is excluded.”

Highlighting a research bottleneck

Professor of Neuropathology, University of Oxford; Former Director, UK Brain Bank for Autism and Related Developmental Research

Still early: “This is an interesting paper but one whose significance is hard to assess. The technique is a relatively novel one that has been used very little so far on human postmortem material. Although the researchers have made efforts to ensure that their results are not influenced by the nature of the material studied, it will take more research before it’s clear just what this study is telling us.”

It takes brains: “There is a major difficulty in having this study independently replicated because there is a shortage of postmortem brains donated for research from young people with autism or controls. There needs to be more awareness of the tremendous value that postmortem brain donation has for research. Studies of the brain based on imaging in life have produced much important new information. But imaging is unable to take us down to the level of detail that can inform us about patterns of gene expression and protein composition at the cellular and molecular levels.”

Watch this space: “This is the type of information that can lead to new understanding, and that has the potential to assist in developing interventions to help people with autism and other developmental disorders. For the time being, though, it is a matter of ‘watch this space.'”

About Cross Talk
Discussions among leading experts in the field. Submit your questions to
  • Richard Blaber

    I have Asperger Syndrome and epilepsy, with a left parietal lobe focus. MRI scan shows evidence of a lesion on the left parietal and of a cortical dysplasia. I also have minimal left-sided cerebral hemiatrophy, and EEG shows an abnormal alpha wave rhythm from the left temporal lobe. I’m very interested in what this article has to say, & have already agreed to donate my brain post-mortem to the Oxford Brain Bank for Autism. Btw, can we please be less negative and medical about autism – it isn’t a ‘disorder’, it’s a ‘condition’. We should celebrate & accept neurodiversity. No autists, no Mozarts, no Einsteins! Beside all the negatives, I have a verbal IQ score of 150, a photographic memory, and am educated to PhD level!

  • Amanda

    This is a good discussion. The Courchesne paper was over hyped. We need a thoughtful discussion of these issues and an examination of pre and postnatal factors.

  • Eric Courchesne

    Reply to Comments by Flora Vaccarino
    During fetal cortical development, programs of interacting genes underlie the processes that generate cortical layer-specific neuron types, correct numbers of these different types, correct layer positioning of each type, and typical neuronal sizes, shapes and connectivity patterns. In our NEJM 2014 study, we found patches of cortex in which neuron-specific and layer-specific patterns were disrupted. In another paper, we found abnormal neuron numbers (Courchesne et al, JAMA 2011) and in a third paper we found genetic evidence implicating cell cycle networks, cell fate and differentiation processes, DNA damage response functions, and cortical patterning (Chow et al., PLoS Genetics 2012). Thus, it appears that key programs of interacting genes that regulate multiple early neural developmental events are disrupted. In forthcoming papers we have additional evidence about genetic programs that may be disrupted and disturb early brain growth in autism.

    Flora Vaccarino notes there are area-specific cortical disruptions of gene expression in autism. In Chow et al, PLoS Genetics we reported downregulation of gene expression for several cortical patterning genes (NODAL, HOXD1, FGF1, NDE1, PCSK6) in prefrontal cortex in young autistic cases; this evidence points to prenatal alterations in cortical patterning in autism. Such data as well as our new laminar disorganization evidence points to disruption of higher-order processes, and not just isolated single genes.

    In our NEJM report we saw evidence of disrupted layers in the focal patches of cortex and, in some instances, there were macroscopic alterations of cortical surface structure (e.g., see apparent instance of polymicrogyria in our Figure 1) and nearby clusters of what appeared to us to be mis-migrated cells. Neuropath examination also noted clusters of disorganized cells. In Figure 1, near the patch region shown, there was a cluster of what appeared to be mis-migrated RORB cells. We suspect that with a larger set of markers and larger blocks of cortex than were available to our study, the presence of laminar anomalies and disorganization, cell differentiation abnormalities, and migration defects might be even more evident.

  • RA Jensen

    Most cases of 22q11.2 deletion syndrome are not inherited. The deletion occurs most often as a random event during the formation of reproductive cells (eggs or sperm) or in early fetal development.

    One can look at mouse models for additional insight. Meecham et al (2009) autopsied the brain of mice with the 22q11 deletion. Diminished 22q11 gene dosage disrupts cortical neurogenesis and interneuron migration. Such developmental disruption may alter cortical circuitry and establish vulnerability for developmental disorders, including schizophrenia and autism.

  • biostatistician

    I just read Chow et al. The evidence for their claims are either weak or obfuscated. Many of the reported findings in that paper are not corrected for multiple comparisons (most are at p < 0.05). For example, in Table S2, if one takes 20,000 probes and asks for differential expression at p < 0.05, one expects 1,000 false positives. The authors found 106 probes differentially expressed at that threshold - they are likely all false positives. For selected analyses, corrections for multiple comparisons are reported using the false discovery rate - the FDR. However, they are reported with varying unconventional thresholds (FDR < 0.1, FDR < 0.27) and sometimes not reported at all. This is a hallmark of reporting only what is necessary to fit preconceived notions. One sees this a lot in the neuroimaging literature. These issues make the claims in Chow et al. difficult to interpret from a statistical perspective. Additionally, Chow et al. is also based on a very small sample size. Until replication of their data is shown (ideally by an independent research group) and stronger data and analyses are provided, I would caution over interpreting their findings. You may be reading signal from noise.


Log in to your Spectrum Wiki account

Email Address:



Request your Spectrum Wiki account

Spectrum Wiki is a community of researchers affiliated with an academic or research institutions. To be considered for participation, please fill out this form and a member of our team will respond to your request.


Email Address:

Title and Lab:

Area of Expertise: