Skip navigation

Calls to Action

Autism Research Progress Reported at the 2007 Society for Neuroscience Annual Meeting

November 29, 2007

Each year scientists interested in the workings of the brain gather at the Society for Neuroscience conference. The Society for Neuroscience meeting is one of the largest gatherings of scientists in the world and provides the focal point for what's new in neuroscience research.

This year, just a week after devastating fires hit much of Southern California, the city of San Diego managed to welcome more than 31,000 attendees from around the globe in the San Diego Convention Center.

Although the pleasant and relaxed surroundings provided a loose and comfortable environment, the conference showed the amazing power of a large scientific community when focused on problems such as autism: as with each of the past three years, the number of presentations concerning Autism Spectrum Disorder (ASD) increased again – this year to over 200. Two entire poster sessions (totaling 39 presentations) were solely dedicated to autism, while another 4 sessions looked at autism research in combination with related developmental and neurological disorders. Over 80 other sessions had interspersed lectures and presentations with autism-related topics. The presentations covered all fields of neuroscience research in autism, ranging from neuroimaging to neuroimmunology to neuropsychology.

Understanding Brain Circuitry in Autism

If one would seek an overarching theme of this year's huge conference, certainly the desire for the understanding of neuronal circuits in the brain would come to mind. The immensely complex neuronal highways that interconnect one brain area with another, as well as the shorter but incredibly dense connections between neighboring neuron populations, and the dynamic strengthening and weakening of these connections, were all subject to investigation. Knowing how the brain can wire and re-wire itself and how these connections differ in people with neurological problems such as ASD, could go a long way in helping with understanding and ultimately alleviating these problems.

On the first full day of the conference, Sebastian Seung Ph.D. of the Massachusetts Institute of Technology gave an inspiring lecture about attempting to map all the connections in the human brain to create a "wiring diagram." He compared the task to the human genome project, which just a few decades ago seemed like an insurmountable undertaking. Currently the only organism for which we know all the neuronal connections is the worm C. elegans, often used as a model system to study neurological problems for precisely this reason. Yet the worm only has 300 neurons with 7,000 connections, while a human brain has approximately 100 billion neurons with 100 trillion connections -- that's a one with 14 zeros! To start this tremendous task Dr. Seung collaborates with Dr. Winfried Deng in Heidelberg, Germany to employ electron microscopic tracing as well as machine learning techniques, terming the process "connectomics."

While Dr. Seung and his colleagues begin the daunting process of tackling the circuitry of the whole human brain, other researchers at the meeting reported that progress has been made in understanding the differences in known neuronal networks affected in autism. One such network is the mirror neuron system, which appears to be important for imitation. Several labs have reported that mirror neuron activity in individuals with autism appears to be lacking. For example, by measuring brain waves, the laboratory of Jaime Pineda, Ph.D. at UC San Diego had previously found that children with ASD are deficient in activating this network when observing the actions of others. However, new research from the Pineda lab now shows that the mirror neuron system is active when children with ASD observe the actions of people they are highly familiar with, and the level of activity is comparable to neuro-typical children. This exciting news means that individuals with autism are not completely lacking a mirror neuron system, but may instead be unable to activate it under certain circumstances. If this holds true, treatments designed to normalize mirror neuron activity become conceptually possible. Indeed, using a grant from Cure Autism Now/Autism Speaks, Dr. Pineda and colleagues are currently exploring whether neurofeedback can be used to "train" mirror neurons in children with autism.

By combining previously separate lines of research, scientists at the conference also reported making progress in understanding how the circuits in brains of individuals with autism may differ molecularly. Serotonin, a neuro-modulator thought to be important for social interactions, has been studied extensively in autism. Serotonin levels in autism have been reported to be abnormal both in the blood and the brain, and genetic studies have pointed to a possible abnormality in serotonin-related genes as well. Independently, neuroimaging studies that searched for regions of the brain that underlie social behavior have indicated that posterior cingulate cortex modulates social and emotional behavior. Now the group of Gene Blatt, Ph.D. at Boston University has used postmortem tissue from the Autism Speaks Autism Tissue Program to show a decreased density of a certain serotonin receptor subtype (5HT2a) in all layers of the posterior cingulate cortex from individuals with autism. Dr. Blatt explained that this could point to a possible explanation for why selective serotonin re-uptake inhibitors (SSRIs) have shown a positive effect in the treatment of people with autism. (Read more about this serotonin study in the Autism Tissue Program companion article)

Progress in Generating Models for Autism

Animal models have long been employed to replicate some of the behavioral characteristics of Autism Spectrum Disorder. However, only recently with the progress of genetic studies in developmental disorders have these models been based on the actual genetic differences seen in humans with autism. Several groups at this meeting reported engineering mice to carry genetic variants that have very recently been associated with autism, such as a mouse that carries a neuroligin 3 mutation and a mouse that carries a variant of the serotonin transporter gene found more often in individuals with autism. The researchers are now testing these animals for signs of autism-like behaviors and, excitingly, reported future plans to use these mice to search for treatments.

One surprising example was shown for mice with mutations in the MeCP2 gene, the gene indicated in the developmental disorder Rett syndrome, which shows some overlap with autism. These mice have many of the same behavioral characteristics as humans with Rett syndrome (including some features of autism), and such studies have now made possible experiments addressing how to treat these symptoms. In very exciting preliminary data, research groups in Italy and at Wellesley College have independently found that mice with MeCP2 mutations that grow up in an "enriched" environment, such as extra-large home cages with toys, exercise wheels and groups of other mice, have improvements in motor coordination, structural changes in their brains, and show a life span that is closer to normal than that of MeCP2 mutant mice that grow up in a standard "unenriched" environment. Although the researchers don't yet know how to translate these results to humans, it suggests that neurodevelopmental disorders caused by genetic mutations are indeed accessible to early interventions (and ones that need not necessarily involve fixing the underlying genetic abnormality). Such studies highlight the importance of creating and studying model systems for autism.

Models of Environmental Agents and Autism

Other groups presenting data at the conference made use of animal models to study the possible role of environmental agents in triggering autism, especially exposure during critical periods in development. Posters presented at the meeting showed that teratogens, such as ethanol or the anti-convulsant drug Valproic Acid (VPA), and substances that disturb the balance of neuro-modulators, such as SSRIs, can result in some autism-like characteristics in adult animals when given to their mothers during early pregnancy. In addition to behavior, researchers are now studying the brains of these animals to understand the changes that result from the short exposures to these substances. Henry Markram, Ph.D. at the Brain Mind Institute in Lausanne, Switzerland has used the VPA model to investigate changes in cortical connectivity and plasticity that may ultimately be important for understanding symptoms of ASD. Because of their data, Dr. Markham and his colleagues currently theorize that the brains from animals exposed to VPA during pregnancy become hyper-excitable.

Looking into other environmental agents that can cause ASD-like features, researchers presented a novel model utilizing the short fatty acid Propionic Acid (PPA). The effects of PPA are particularly intriguing because it is a common by-product of a subpopulation of human gut enterobacteria and is a common food preservative. Researchers at the University of Western Ontario, Canada injected PPA directly into the brains of adult rats and found social, cognitive, and sensorimotor impairments consistent with human ASD. With single injections the effects were temporary (30 min), but repeated injections induced diffuse neuroinflammation and changes in the blood-brain barrier, which the researchers report may be similar to those found in people with ASD.

The New Immune System and Autism

Another potential environmental influence on ASD that is garnering a lot of attention is the role of the immune system. Several posters reported animal models of prenatal immune activation in pregnant mothers and the effects on their offspring.

In a follow up to a 2005 study, new preliminary data reported at this meeting, and funded by Cure Autism Now/Autism Speaks, showed evidence of immune activation in postmortem brain tissue from children with autism, even at very young ages. (Read more about this immune activation study in the Autism Tissue Program companion article)

Others at the meeting reported initial findings of abnormal immune gene expression in subgroups of people with autism. (Read more about this gene expression study in the Autism Tissue Program companion article)

Perhaps the most interesting session related to immune function was organized by Cure Autism Now/Autism Speaks grantee Lisa Boulanger, Ph.D. from UC San Diego. The symposium was titled "The New Neuroimmunology: Immune Proteins in Synapse Formation, Plasticity, and Repair." The speakers in this symposium addressed the possibility that molecules previously thought to function strictly in the immune system may actually also have important roles in brain development. Although the direct relevance to autism remains a hypothesis, Dr. Boulanger told the packed audience that, rather than the brain influencing the immune system (such as with stress), or the immune system influencing the brain (such as with sickness-induced lethargy), in the new model, the immune and nervous systems could share similar molecular features. In such a model, something that affects one of the systems may also affect the other in parallel. (For more information on this special session, please see In a later session, Cure Autism Now/Autism Speaks grantee Kimberly McAllister, Ph.D. from UC Davis presented new data suggesting that immune proteins also may be serving in the brain to actually inhibit the formation of synapses, the structures that allow nerve cells to communicate with each other.

Improving Autism Symptoms

A final environmental influence, but one that actually may help individuals with ASD, is a good night of sleep. In research sponsored by NAAR/Autism Speaks, Beth Malow, M.D. and her colleagues at Vanderbilt University have found that children (age 4-9 years) with ASD who have low levels of melatonin also have decreased levels of deep sleep. While more research is required before doctors can recommend melatonin supplements, which are easily available over the counter, Dr. Malow says that research correlating deeper sleep in children with ASD who have greater levels of natural melatonin is very promising (for further information on this study, please see

Other encouraging results came from researchers at Carnegie Mellon University in Pittsburgh on the final day of the meeting. They investigated brain activation during a task that measures executive function in the prefrontal brain regions. While overall the researchers found decreased recruitment of important brain areas in people with autism, the researchers uncovered a continued improvement from adolescence to adulthood in the individuals with autism, showing that the window of plasticity and circuitry improvement found during typical development appears to also be available to people with autism. This provides extremely encouraging news and additional hope for the remediation of autism -- even into adulthood.

It has to be emphasized that almost all the research described above is preliminary and still needs to be confirmed, published and replicated. However, the enthusiasm inspired by this year's Society for Neuroscience conference lingers on and, if the progress in autism research made in the past year is any indication of what is going to happen in the coming years, the excitement provided by much of the early data presented at this year's conference may soon translate into hope of real progress in the lives of those afflicted by the disorder.

(Walter Lerchner, Ph.D and Sophia Colamarino, Ph.D., with additional reporting by Scot Thomas, M.D.)