A new Autism Speaks-funded study reveals differences in brain connections in areas that redirect attention, mediate social interactions and modulate emotional responses.
Science is one step closer to understanding how the brains of individuals with autism process information. Over the past decade it became clear that neural communication is disrupted in autism, although the details, and the causes, remained elusive. Two hypotheses emerged to explain the information processing differences found in individuals with autism spectrum disorder (ASD). First, scientists identified over-activity at the level of the synapse—the point where two neurons connect. Second, imaging studies showed that the balance of local versus long-range connections was skewed in autism in favor of strong local connections but weak long-range connections between distant brain regions. Complex behavior, such as language and social interaction, depend upon such long range connections. Although the findings supporting the two hypotheses were likely related, the mechanism relating over-activity at the synapse and biased local versus long-range connectivity was unclear.
Now, taking advantage of recent technical developments that facilitate the measurement of small structures within the brain, Autism Speaks’-funded researchers at Boston University have bridged these two hypotheses. Basilis Zikopoulos, Ph.D., and Helen Barbas, Ph.D., reconstructed the intricate fabric of neural “wires”—called axons—that connect prefrontal cortex with neighboring and distant brain regions. The prefrontal cortex is a collection of brain areas that are involved in higher cognitive functions such as attention, planning, social interactions, and abstract thinking. Using postmortem brain tissue generously donated by families of individuals with and without autism, the team measured features of the different axons traveling beneath the cortical surface.
Talking to itself
In comparison to the control samples, autism brain tissue had fewer large axons connecting regions of the prefrontal cortex to the other areas of cortex. Added to this connection imbalance is a thinner coat of axon insulation, called myelin. Myelin is a fatty substance that insulates axons to ensure faithful delivery of neural signals over long distances. Prefrontal axons from autism brain tissue traveling to distant brain areas have less myelin. Zikopoulos and Barbas also found more thin axons that connect neighboring regions in the prefrontal cortex in the autism brain tissue versus the control samples. This biased pattern of connectivity creates a situation in which prefrontal cortex is not efficiently exchanging information with other brain areas, but rather is operating in a solo fashion.
Pursuing the reason for the abundance of thin axons, the researchers first checked to see whether this was due to more neurons in the cortex just above. Detailed examination of neuron density showed no more neurons, just more axons. This finding led the team to speculate that atypical axon branching had occurred in the autism samples. By painstakingly tracing individual axon branches through a volume of tissue, Zikopoulos and Barbas confirmed that the abundance of thin axons emerged from overzealous branching. The thin axons found in autism samples formed tree-like braches that connected with many more cells in neighboring areas than is typical. This pattern of branching furthered the bias for local versus long-range communication.
Differences in axons and their insulation may offer some links between the two hypotheses that aim to explain disrupted communication in autism. Fewer large axons travel to distant brain regions. Thin axons form too many branches that make extra connections locally. These factors can handily explain the local over-connectivity and long-range under-connectivity observation made through imaging studies. The fact that small axons seem to branch more frequently would also lead to an over-abundance of synaptic connections in a neighboring region. This result can help to explain the synaptic connection hypothesis. Taken together, the results lead us to a deeper understanding of how information may be processed quite differently in the brains of individuals with autism.
Anatomy offers a closer look
Imaging studies aiming to understand brain connectivity use techniques such as magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI). These techniques focus on the large swaths of axons, often called “white matter” because of myelin’s fatty white hue. While many of these studies have suggested an imbalance favoring local connectivity in the brains of individuals with autism and served as a stimulus for this study, the level of detail that comes from quantitative anatomy is much richer. “The study of brain tissue at high resolution is needed to pinpoint defects in the brain’s communication system, which is prerequisite to developing rational therapies. Studying pathways using DTI is comparable to looking down from a plane at 35,000 feet – you can see some of the major highways below but not the exits, and certainly not the driveway to your house to see if the car is there,” states Dr. Barbas.
Martha Herbert, M.D., Ph.D., of Massachusetts General Hospital and LADDERS, an imaging expert who was not involved with this study, was enthusiastic about the new results. “Zikopoulos and Barbas have given us a paper about the cellular level of white matter changes in autism that has needed to be written for quite some time. Their analysis is meticulous and comprehensive, and goes far in providing microscopic details to answer provocative questions raised by the larger scale observations in so many brain scan studies in autism. Their findings suggest that the insult occurs later in brain development, after the neurons migrate to where they will live in the brain. The time when these neurons hook up with each other through white matter is crucial and during this period environment and immune factors as well as genes may play a role in the observed changes. These specific details about white matter in autism matter greatly for how we understand autism and how we go about providing the help people with autism need.”
Modeling the development of axon differences
The new axon data may also offer suggestions of what occurred to lead to this state of altered connectivity. Combining the axon data with what is known about the development of the cerebral cortex, the researchers devised a model that links differences in axons below frontal areas to a cascade of events in development that affects the growth of axons and their insulation. The hypotheses proposed are testable, and point to specific pathways that could to be targeted for the development of therapeutic interventions in autism.
One pathway that Zikopoulos and Barbas analyzed in detail was GAP-43, one of a myriad of signals that are exquisitely timed to encourage the proper growth and integration of developing neurons. GAP-43 stimulates axon branching. GAP-43 also interacts with other signals to prevent premature myelination, and is turned down after myelination begins. In the anterior cingulate cortex, a region of the prefrontal cortex that was the most profoundly affected in the brains studied, the neurons are sensitive to GAP-43 signals for an especially long duration in development. This is an important point because environmental agents such as estrogen-like compounds can increase the GAP-43 signal altering the timing of different aspects of neural development.
One take-home message from this study is how important it is to have diverse lines of inquiry investigating autism. The axon results were consistent across individuals studied. For example, the same changes were found regardless of whether or not the individual experienced seizures. Zikopoulos and Barbas also note that many of the genes that have been associated with autism exert effects on neural connectivity. Dr. Barbas says, “Genome-wide association studies are valuable in pointing out potential vulnerability in affected individuals. Our study suggests further that environmental factors (in utero, after birth, or both) can lead to abnormal connectivity because external factors up-regulate signals such as GAP-43 that regulate axon growth. We need studies to take into account all of these aspects of autism, not just one.”
This work was published in the Journal of Neuroscience and featured on the cover of the November 3 issue.
This research was made possible through funding from Autism Speaks and brain tissue supplied by Autism Speaks’ Autism Tissue Program. These groundbreaking results came from the brain tissue of ten generous individuals who chose to make a lasting gift to science. Importantly, only a small portion of each donated brain was needed for this study, making other sections of this incredibly valuable research resource available to others to uncover new insights into autism spectrum disorders. Please consider making a lasting gift to science by participating in the ATP.