Several years ago my good friend Eric London (see bit.ly/1n6uqc7) wrote the attached letter when I resigned my position as member of the Scientific Advisory Board for NAAR. Back then members of the Scientific Advisory Board were prohibited from pursuing any research into autism in order to avoid any conflict of interest when judging grants for possible funding. As it would happen, shortly after my resignation my first grandchild was born and we soon suspected that something was wrong with him. Although fulfilling the criteria for autism, he was severely impaired and many physicians suspected a genetic defect (i.e., syndromic autism). He never developed language and has suffered from gross motor deficits, extreme sensory perception (touch) and relentless seizures. We took Bertrand to many academic institutions and were given various tentative diagnoses (Rett syndrome, Dravet) but little in terms of possible treatments. Interventions, if anything, brought serious side effects. In the end, a complete sequencing of his genome showed a mutation in a gene (NGLY1) and he became the first subject ever described with this mutation. It is curious how life placed me on this pathway. Trying to help close friends of mine who had autistic children, only to find out all along that I would be trying to help my own family. Karma.

Casanova: Draft Naarrative Article Summer 2003
By Eric London, M.D., Vice President-Medical Affairs

For anyone who cared about autism spectrum disorders, it was a frustrating experience attending scientific meetings in 1996. So many disciplines in the neurosciences seemed to be taking off with all sorts of monumental accomplishments. Reading the scientific literature often seemed like science fiction with everything changing so rapidly. At the same time, there was very little research pertaining to autism. In response to this dire situation, one of our first and pivotal missions at NAAR was to attract accomplished scientists who had never studied autism and do our best to persuade them to change their research focus and dedicate themselves to autism.

I have been pleasantly surprised over the years as to how many scientists actually have became interested in autism and changed the course of their own research careers. We are proud to say that several of these scientists have come from NAAR’s own scientific advisory board. They were originally recruited for their expertise in areas unrelated to autism and, over their tenure serving on NAAR’s all-volunteer Scientific Advisory Board, became interested in our children’s plight. One of these scientists is Dr. Manuel Casanova of [Georgia State University School of Medicine].

We first were introduced to Dr. Casanova when NAAR was seeking a scientific advisor on our Scientific Advisory Board knowledgeable about researching human brain tissue. Dr. Casanova is considered one of the world’s experts on the technical aspects of how to handle brain tissue. In fact, he was instrumental in the organization of our now very successful Autism Tissue Program. His own research focus was schizophrenia. Despite having no personal connection with autism, he completely embraced the task of serving our organization in any way that he could. He always made himself available for the numerous tasks that NAAR called upon him to do. Despite the personal sacrifice that he made in time and energy to help us, I was always surprised—and touched– by his expression of gratitude to us for the efforts we made in supporting autism research. Over time, Dr. Casanova’s compassion and incredible dedication to NAAR’s mission became clearer. For I learned that Dr. Casanova is also the parent of a child who faced a serious medical problem. He too undertook every effort—as father and scientist– to help his child.

Dr. Casanova is now doing some of the most exciting and novel work in all of autism research. I would now like to briefly describe his research and discuss some of the implications that I believe it has for the future of autism treatment and prevention. In order to appreciate his work, it is first important to describe briefly what we currently know about neuroanatomy in general and in individuals with autism.

In truth, brain sciences or the neurosciences are still in their infancy. While other organs of the body have been studied systematically for many decades, it is only a recent phenomenon that we are able to study the human brain. The main reason for this is the relative inaccessibility of the human brain. Ironically, in this era of high tech medical procedures, we are first learning how to navigate around the brain’s bony protection that is, of course, the skull. Another reason for our lack of knowledge about the brain is its delicacy. For example, when doctors believe that there is pathology in the liver, they can quite easily secure a biopsy and directly examine the tissue. Brain biopsy, although possible, is reserved for the most serious problems. As late as the 1970’s, the only “safe” way to look at the brain was an x-ray that yielded almost no information. Since then, a succession of new techniques has become available, starting with the CT scan, followed by the PET scan and the MRI. As these techniques evolved, we can now study the structure– and even the functioning– of the brain safely and at reasonable cost.

Another method of studying the brain, postmortem examination, has been available for a long time. Although available for research, the methods used were relatively crude and full of technical obstacles. For example, the time between death and preserving the tissue can lead to massive changes making the tissue virtually useless for study in some circumstances. The sheer number of brain cells and the complexity of their connections were impossible to examine until recently. Today, with the aid of high-speed computers and scanning devices, much more sophisticated methods are available to study neuroanatomy.

An organ such as the liver is comprised of only a few types of cells. Sixty percent are hepatocytes and they account for 90% of the volume of the organ. Therefore, the liver is largely a homogeneous organ. Studying any part of the liver is very similar to studying any other part. This is not true of the brain. Although it is said that the “neuron” is the fundamental type of cell involved in brain functioning, in reality there are many types of neurons. These cells have been classified into different types using the different neurochemicals that they use to communicate (such as serotonin, dopamine, etc.). Brain cells also grow connections called dendrites that enable them to form efficient connections to other cells. These connections go on to form pathways through the connection of several cells, sometimes called a circuit. The complexity is increased further by the reality that each cell may communicate with a huge number of other cells so that the pathways cannot be considered closed circuits. Cells can enhance, or inhibit, the functioning of a given circuit. But the real complexity is not understood unless it is understood that there are literally __ brain cells with __ numbers of connections or synapses.

With this backdrop, let us consider what is known about the brain in individuals with autism. Among all the biologic abnormalities identified thus far in autism, the current evidence for anatomical abnormalities is actually the strongest. Some of these findings have been highly reproducible and are seen in both scanning and brain tissue examination. There is evidence of various brain structures being abnormal. Despite this evidence, what we really know is generally restricted to merely labeling a structure in the brain as being either too large or too small. We have some information about abnormal numbers of cells in various areas and some suggestions of abnormal distribution of cells but, in reality, there is scant evidence of what actually is “wrong” in the autistic brain.

In fact, until we understand more about the differences between the brains of individuals with autism and the brains of neurotypical individuals, I believe that we will continue to be thwarted in our desire to know “what went wrong.” As a result, the likelihood of securing meaningful treatments will also remain remote. However, the situation is clearly improving with new and more powerful neuroimaging techniques becoming available and enabling us to learn far more about autism. The availability of postmortem tissue through the Autism Tissue Program has made an enormous impact–enabling a total of 30 investigators to undertake brain research in autism from the perspective of neuroanatomy as well as neurochemistry and genetics.

The most compelling evidence of anatomical abnormalities in individuals with autism has been described in the lower portions of the brain; that is, the brain stem and the cerebellum. There is also some evidence for abnormalities in other structures such as the amygdala and the hippocampus. An extremely important part of the brain (and the largest) is the cerebral cortex. Although many of the symptoms noted in autism seem to be functions of the cerebral cortex, neuroanatomically it has been difficult to document any abnormalities.

Dr. Casanova has spent many years of his career studying a tiny structure found in the cerebral cortex called the “minicolumn.” These structures consist of 80-100 neurons arranged radially like pearls on a string and they are found in all areas of the cortex. Minicolumns are believed to comprise the smallest level of functional organization in the cortex. There also exists a structure called the “macrocolumn” that consists of 60-80 minicolumns. These structures are very hard to “see.” That is, these are normal cells that line up in space in such a way as to enable them to function together. Imagine flying over a farm. Suppose the farmer planted his tomatoes in clusters with just a little extra space between the clumps. Unless one were looking for this arrangement, it might not be obvious. In addition, it might only be noticed if one were looking at just the right angle. This could be an explanation why minicolumns have received little attention until now.

Each minicolumn might function as a “processing unit.” That is, each individual brain cell in the cortex is itself probably incapable of performing meaningful work in such a complicated system. Rather, the brain cells perform work within the minicolumn that then receives and sends processed messages. [They] receive messages from distant parts of the brain but also communicate with other minicolumns in the cortex. A key to understanding how minicolumns function is to understand that there is a sophisticated group of “inhibitory circuits” within them. Different types of inhibitory fibers modulate the functioning of the minicolumns in different and very subtle ways and, therefore, are able to undertake the sophisticated processing of which the brain is capable.

All of this was quite academic to those of us interested in autism until a paper published in the prestigious journal, Neurology, last year. In that article, Dr. Casanova and his colleagues reported abnormalities in the structure of minicolumns in the brains of autistic individuals. Specifically, they found that, in the brains of individuals with autism, the minicolumns were more numerous but smaller than in “controls” (individuals unaffected by disease). They also found that there was less space between the minicolumns in the autistic brains. Because minicolumns cannot be visualized on MRI scans, this work has been done on post mortem brain tissue. Dr. Casanova and his colleagues reported on nine subjects with autism and four controls. In a separate publication, they reported on one brain of an individual with Asperger’s Syndrome with similar results. Although this research team has found abnormalities in other brain diseases and disorders, such as schizophrenia and learning disabilities, the nature of the abnormalities were completely different and, in fact, more dramatic in autism.

As in all scientific studies, Dr. Casanova’s research findings will require further study and replication. However, if these findings are substantiated, what would they indicate about autistic disorder? First, from a point of view of brain development, a key question is why are the minicolumns abnormal? At this point, it is not known what would cause this. It is believed that minicolumns are formed by both genetic as well as environmental influences. By studying the factors that form the minicolumn, it would be possible to model the factors that may be causing the differences found in the autistic brains. It should be noted that minicolumns are found in all primates, so animal models could be created with direct bearing on autism.

Another and perhaps more immediately testable clue could come in explaining how the brain functions in autism. For example, it is widely observed that individuals with autism have “processing problems,” such as auditory processing or what is commonly called sensory integration problems. At this time, when we talk about “processing” or “integration”, we really do not know what that means on a cellular level. Dr. Casanova’s neuroanatomic findings could provide a model by which this could be explained.

One model for how the brain malfunctions in autism has been described as the “temporal binding deficit hypothesis of autism.” Whereas in typical development the brain develops specializations but also “integrates” the different functions, in autism the functionally specialized brain regions become increasingly isolated from each other. Those cognitive faculties that require the integration of numerous brain regions do not develop while those that rely on more localized neuronal activity can function well. This could explain the existence of splinter skills– and even the development of extraordinary talents known as savant skills– while at the same time leaving the affected individuals with significant problems in generalizing what is learned.

Recent research has also focused on the electrical activity measured on EEG in the frequency known as “gamma.” It seems that groups of cells firing at about 40 Hz results in an oscillatory pattern are able to communicate their functioning to other remote areas of the brain and make that area fire at that same frequency. It could be hypothesized that, in autism, the brain is unable to communicate between brain regions because of an inability of some areas to fire at the correct frequency. This phenomenon has been labeled “hypocoupling,” leading to the state in which local networks of cells do not correlate temporally with one another. In order for proper “coupling” to take place, the timing has to be just right.

Dr. Casanova speculates that these electrophysiological phenomena can be explained by the abnormalities in the minicolumns that he has observed in individuals with autism. Because of the finding of smaller and too numerous minicolumns, it could be that the cortex in autism is firing too many processing units at once, which are unable to correlate with each other to create a sufficiently coherent response to produce the correct oscillatory cycles. To produce these cycles, one needs thousands of cells to fire in a coherent fashion.

This theory could lead to direct and immediate clinically relevant information. We know that seizures are caused by abnormal electrical activity. In seizure disorder, if too many cells (as opposed to just the right amount of cells) fire in synchrony the brain is overwhelmed and “shorts out,” to use the electronics metaphor. Seizure disorder is very common in individuals with autism. Even more common is “epileptoid activity“ seen on the EEG, which are seizure-like findings in the absence of actual clinical seizures. The abnormalities in the minicolumns that Dr. Casanova identified offer an attractive new hypothesis as a cause of seizure activity in individuals with autism.

As those of us familiar with autistic children know well, there also seems to be a rather narrow “window” of external stimulation during which a child can learn. Many autistic children seem not to respond to low levels of stimuli. It is often difficult to secure their attention without specifically calling for it, such as saying, “look at me” or even at times directing their faces. At the same time, many individuals with autism have a very difficult time focusing when there is too much stimulation, such as in a noisy classroom. They are unable to stay focused and often get agitated. A unifying explanation of these observations could be that the individual needs just the right amount of stimulation in order for the brain to process information efficiently. This observation has led those who practice applied behavior analysis to orchestrate classroom conditions so as to maximize the opportunity for learning or, put another way, to target just the right amount of “arousal.”

This arousal argument might also be explained by Dr. Casanova’s minicolumn findings. If the processing units are not well synchronized, then over or under arousal could easily occur.

One hopeful outcome of this research is the possibility of using pharmaceutical interventions to regulate the function of the minicolumns. As stated above, we know that there are GABAergic inhibitory fibers within the minicolumns. If the firing of these units could be regulated, then the brain’s ability to learn and the increased seizure activity may be more effectively treated.

All of us at NAAR have been excited by Dr. Casanova’s recent involvement in autism research and by his intriguing findings. We were saddened– but also heartened– by his decision last fall to resign from NAAR’s Scientific Advisory Board after six years of volunteer service to our organization. We were saddened since we understood that NAAR has lost a great asset to its evaluation process. However, we were also heartened be the reason why he resigned. Dr. Casanova indicated a desire to apply to NAAR for funding to further pursue his autism research and he knew that NAAR maintained a policy precluding it from funding its own advisors. As such, Dr. Casanova’s resignation from NAAR’s Scientific Advisory Board means one less distinguished advisor for NAAR but one additional distinguished scientist devoting his efforts to the autism community.

There are never enough occasions to express our profound gratitude to NAAR’s Scientific Advisory Board for their enormous commitment of time, energy and talent to our loved ones and our cause. We wish to express them today to Dr. Casanova.