For the second year in a row autism was featured at the United Mitochondrial Disease Foundation meeting. Following last year’s well-attended afternoon symposium, Robert Naviaux, M.D., Ph.D. (UCSD), in conjunction with Autism Speaks’ science team, successfully applied for an NIH conference grant to support a more extensive full-day meeting that included a family “Ask the Doc” panel discussion.
Mitochondria are the primary energy factories for all cells in the body. When these factories reduce their output, critical cell functions begin to flicker or fade. The energy is produced through a process called oxidative phosphorylation—an elaborate process that converts oxygen in body tissues to energy used for all cell functions. Although the metaphor of “energy factory” is the most common way to think of mitochondrial function, mitochondria are responsible for much more. Mitochondria were once — long ago in the history of evolution — single-celled organisms like bacteria that functioned independently, responding to the environment and producing their own proteins encoded by a small circle of DNA. Several billion years later, mitochondria are fully integrated into our cells, co-opting proteins encoded by the much larger nuclear genome of each cell to serve different functions. Proper mitochondrial function is tuned for each cell type since skin cells, heart cells, and brain cells all have different energy needs. Mitochondria, like their bacterial predecessors, remain exquisitely sensitive to the local environment and as such function differently in developing versus mature cells, and also in response to differing temperatures, toxins and immune challenges.
Genetics and Epilepsy: finding common ground
The Saturday scientific session began with two primers, one on Autism Spectrum Disorders from Sarah Spence, M.D., Ph.D. (NIMH) and one on mitochondrial disorders from Salvatore DiMauro, M.D. (Columbia University). This background allowed the audience of parents, researchers and clinicians with different specialties to find some common ground for the later presentations and discussion.
One of the unique aspects of this meeting was the pairing of two talks on a topic, one taking the perspective of autism and the other mitochondrial disorders. Abha Gupta, M.D., Ph.D. (Yale) informed the group about our current understanding of autism genetics and focused specifically on what can be learned about the biology of autism by discovering rare mutations. Dr. Naviaux then wove the perspective of traditional mitochondrial genetics into a broader tapestry for the audience to consider. Over 1100 proteins are active in the mitochondria, and the DNA that codes for these proteins is scattered throughout the mitochondrial genome and all the chromosomes in the nuclear genome. This scatter and spread makes mitochondrial function an easy target for random mutations. Also, mutations in one gene can have complex effects on the expression of other genes.
Another area of Dr. Naviaux’s research has focused on a mitochondrial disease called Alper’s syndrome, in which the patient develops typically until a relatively mild viral infection stresses the system and uncovers the mitochondrial deficit, resulting in the onset of severe symptoms. Research into genetic vulnerabilities revealed by environmental stressors is relevant to our understanding in autism of the interaction of genes and the environment. In his presentation, Carlos Pardo, M.D., (Johns Hopkins University) considered the interaction with the immune system by focusing on how components of the immune system serve key roles in development. Dr. Pardo surprised the autism research world in 2005 when he showed clear marks of neuroinflammation—signs of an activated immune system in the brain—using postmortem tissue from individuals with autism. This intersection of immunology and the brain has been a major focus of his lab. At the symposium, Dr. Pardo showed how a specific class of receptors that are important for marshaling resources to fight infection and healthy brain development also affect mitochondrial function. The converse was also noted– the metabolic breakdown products from dysfunctional mitochondria can adversely affect this unique receptor system.
A second set of presentations focused on epilepsy. Russell Saneto, D.O., Ph.D. (University of Washington) offered his clinical perspective from treating many cases of epilepsy in mitochondrial disorders. Depending on the underlying cause of epilepsy in mitochondrial disease, different types of treatment tend to be more effective. This theme was echoed in a presentation from autism expert Roberto Tuchman, M.D. (University of Miami) who talked about “the epilepsies” as a related group of disorders, but noting that optimal treatment comes from identifying underlying biological conditions when possible. A particular type of epilepsy called West syndrome was described by both speakers. This severe form of epilepsy is thought to involve a dysfunction in a component of neural circuits known as “fast-spiking interneurons” that inhibit the activity of neighboring cells. These fast-spiking cells are particularly expensive from an energetic perspective. Therefore, any mitochondrial dysfunction would especially affect these energy-demanding cells.
Carolyn Schanen, M.D., Ph.D. (University of Delaware) presented data on individuals with autism spectrum disorder that have a duplication in the long arm of chromosome 15. These individuals, who frequently present with epilepsy, exhibit an interesting pattern of gene expression and show evidence of mitochondria dysfunction in postmortem brain tissue. The portrait of this subgroup of autism punctuated the need for pursuing research studies of mitochondrial function in autism and simultaneously highlighting the immediate need for better diagnostic and treatment options.
Diagnosing and treating mitochondrial autism
Richard Haas, M.D. (UCSD) presented the state-of-the-art for diagnosing mitochondrial disorders. The diagnosis of mitochondrial dysfunction is dependent on the results of a series of tests, some of which, like a muscle biopsy or lumbar puncture, are relatively invasive. This presents a situation in which patients and parents need to elect how much testing to do with advice from an informed mitochondrial expert. Although there is no definitive test for mitochondrial disorders, there are agreed-upon checklists based on test results that indicate “likely” or “probably” mitochondrial dysfunction. Dr. Haas pointed to a set of “red flags” that should lead primary care doctors, including pediatricians, to consider a referral to evaluate mitochondrial function.
Treatments for mitochondrial disorders also share a commonality with autism—many complementary and alternative therapies exist with unfortunately little evidence as-of-yet to support their use. Bruce Cohen, D.O., M.D. (Cleveland Clinic) evaluated what we know about vitamin and supplement therapy. Few overall conclusions could be drawn, given the heterogeneity of presentation of mitochondrial dysfunction, but it is clear that more randomized clinical trials are needed especially in subgroups of patients. Until we have more data, exercise is recommended as therapy for all those living with mitochondrial disorders, and certain supplements to support good mitochondrial function and minimize reactive oxygen species may be used under the supervision of a physician to monitor benefit.
Ask the Doc
A group of parents and patients were fortunate to have five leading pediatric neurologists addressing their questions and concerns about mitochondrial disorders and autism. The panelist included Drs. Richard Haas, Sarah Spence, Bruce Cohen, Pauline Filipek, M.D. (University of Texas at Houston) and Roberto Tuchman.
Parents began by asking questions about the prevalence estimates for both autism and mitochondrial disorders, for which we have little population data in the US. However, the panelists explained that data from a large Portuguese study and several smaller studies would suggest that approximately 5-10% of cases of autism may also have a likely mitochondrial dysfunction.
A lot of discussion centered around the utility of genetic testing, with the panel carefully making the distinction between the need to pursue genetic studies for research but taking caution to not put too much weight on genetic studies for individual diagnostics. A comparative genomic hybridization (CGH) array study is definitely recommended as a good place to start for suspected mitochondrial disorders.
Questions surrounding treatment were another hot topic. As noted previously, we would like to get to evidence-based medicine standards for the treatment of autism spectrum and mitochondrial disorders but that is difficult with disorders that present with such unusual heterogeneity. Among the panelists, there was a general consensus that therapies such as chelation and hyperbaric oxygen were not recommended for this population due to a lack of evidence for positive effects paired with substantial evidence for the potential to harm. In particular with hyperbaric oxygen therapy, the panelists were concerned about the potential for reactive oxygen species that can emerge from exposing a weakened mitochondrial system to more of what it can’t process well (that is, turning oxygen into energy). For more innocuous potential therapeutic strategies (such as dietary interventions including some supplements), the panelists suggest that patients (or their parents) work with their physicians to conduct their own trials. There was great hope for pharmacogenomics in that therapies of the future can be targeted to support a known deficit.
The meeting closed on a note of enthusiasm both for this topic and the pervasive sense of collaboration at this meeting. Autism Speaks looks forward to more collaboration between our communities for continued progress in understanding and awareness of mitochondrial disorders in autism.
Staff Blogger, Leanne Chukoskie, Ph.D., Asst. Dir. Science Communication and Special Projects
Epilepsy is a common comorbidity in autism, occurring in as many as 30% of all cases. In overnight or 24 hour EEG studies, atypical brain activity known as spikes or epileptiform activity—that looks somewhat like brain activity during seizure but does not occur with a seizure—has been measured in as many as 60% of autism cases. What can this atypical brain activity tell us about autism? Should we think of the presence of this activity as a biomarker for autism, or is it a more generic sign of brain dysfunction?
At IMFAR, this topic was raised in an invited educational symposium focused specifically on epilepsy and autism. Sarah Spence, M.D., Ph.D., (NIMH; and also serving on the ATN and AGRE board of advisors), a neurologist specializing in both autism and epilepsy, organized and chaired the special session.
The scientific literature indicates that although there is considerable variability in the presentation of seizures among individuals with autism, there are also some consistencies. The rate of seizures in the autism community varies with the age of the individuals studied and the frequency of comorbid symptoms or syndromes and intellectual disability. Although girls are less frequently diagnosed with autism, they are more likely to experience seizures. All types of seizures have been observed in autism, but the generalized tonic clonic (or grand mal) seizure is most common. It was noted, however that it is difficult at times even for a skilled clinician to distinguish what might be absence or complex partial seizures from the behavioral arrest, periods of unresponsiveness, eye deviations and repetitive behaviors that are common in autism.
Michael Chez, M.D. (Sutter Medical Center, UC Davis School of Medicine), offered the perspective of a neurologist who sees many children with autism. Although there are many tests that are ordered for autism, the EEG is the one that most often reveals an atypical result. In a 2006 study Dr. Chez and colleagues reported on 889 children with non-syndromic autism without other comorbidities. Sixty percent of these children showed epileptiform activity even though they had never had a clinical seizure. Importantly, he and others were surprised to see no difference in the incidence of epileptiform activity between children who had regressed and those who showed no developmental regression.
All clinical seizures are treated with medication. But these data about abnormal EEGs in the absence of epilepsy suggest a clinical question: should the epileptiform activity be treated as in the same manner as frank epilepsy? A few recent studies showed benefits not only in the reduction of this aberrant activity on EEG but also some improvements in cognitive function with the use of certain anti-seizure medicines. Randomized controlled trials are needed to learn whether treating epileptiform activity itself offers more potential for improvement than harm.
In some cases of severe intractable epilepsy, steroids have been used to gain control of seizures. This appears to work for a short period of time (long term steroid use has many adverse effects) in individuals who experience seizure activity focused in the right hemisphere along the main fold (called the central sulcus) in the brain.
Jeff Lewine, Ph.D. (MIND Research Network, Albuquerque, NM) and colleagues conducted an open label trial of the anti-seizure drug, Depakote, given with a steroid, and measured how well individuals with autism produce and understand language before and after treatment. Most of the children tested showed substantial improvement in both aspects of language while on treatment. Those who responded to the steroid tended to have simpler patterns of epileptiform activity than those who did not respond to the treatment. Unfortunately, long-term steroid use isn’t a viable treatment option and 80% of the children who showed gains with the steroid lose them again after stopping the drug. A randomized clinical trial of Depakote plus steroid is needed as well as discovery of new methods to induce these improvements without the side effects.
Dr. Lewine also spoke about the typical (but by no means only) pattern of epileptiform activity in individuals with autism, which occurs in both hemispheres and is multi-focal with common activity around the sylvian sulcus. Magnetic resonance imaging (MRI) comparisons of brains of children with epileptiform activity show an increase in brain volume, especially with the white matter or “wires” in the portion of the brain where the activity is focused. The question is which comes first, the increase in white matter or the epileptiform activity? We don’t have an answer to that question yet, but it is an active area of research. We do know that when we consider the incidence of epileptiform activity across the autism spectrum, it appears that individuals with lower levels of functioning have more epileptiform activity. And as noted earlier, the overall incidence of atypical brain activity in children who experienced an autistic regression is not different from what is observed in children with general developmental delay, however some studies show that children who have regressed have greater incidence of epilepsy.
One of the genetic syndromes that presents with a high proportion (approximately 40%) of individuals with autism is Tuberous Sclerosis (TSC). TSC is a rare disorder affecting the brain, the skin and other organs. Individuals with TSC frequently experience seizures (90% likely with a lifetime) whether or not they have autism. The pathological features of TSC in the brain involve the presence of tubers in the cerebral cortex. Within these tubers scientists have observed giant brain cells that are 10 times greater than the size of a typical neuron. This seemingly odd result makes sense with what is known about the effects of the TSC enzyme complex in controlling the growth of cells. TSC acts by putting a break on the cellular machinery that creates new proteins for cell growth.
Dr. Mustafa Sahin, M.D., Ph.D. (Harvard; Children’s Hospital, Boston) presented data on TSC along with an intriguing hypothesis: miswiring of the projections that connect neurons (called axons) results in the start of the disease process in TSC. Animal models of TSC support this hypothesis—TSC mutants have more axons. Scientists examining these animals’ early development also see errors in how axons weave their way through the immature brain to reach the location where they will begin making adult-like connections. Treatment with the drug rapamycin helped animals with dysfunctional TSC genes in terms of improving the insulation on the neural wires (the insulation that covers axons is called myelin), which makes communication more effective in axons that travel between distant brain regions. By improving the process of adding myelin to these animals’ axons, seizures stopped and learning also seemed to improve. Dr. Sahin and colleagues are now beginning Phase II of a randomized controlled trial using rapamycin in patients with TSC to look for neurocognitive improvements with the drug. By learning more about how this drug works in people with TSC, we increase our opportunities for creating therapies for individuals with non-syndromic forms of autism.
Atypical sensory perceptions are commonly experienced in autism and some have noted altered perceptions just prior to the onset of seizures. The immature visual system offers a unique opportunity for understanding how early epileptiform activity in the brain can alter development and perception. Michaela Fagilioni, Ph.D. (Harvard; Children’s Hospital, Boston) and colleagues showed how the outcome of visual development could be controlled in elegant experiments designed to nudge what is known as the “critical period” for determining how much real estate in the visual brain will be dedicated to the right versus left field of vision. The critical period is a time of great adaptability of brain circuits. Administration of drugs, like diazapam (more commonly known as Valium) very early in mouse postnatal development will shift the start and end of the critical period to earlier times. Keeping the animal in the dark and suppressing inhibitory activity in the local brain circuits are two ways Dr. Fagilioni and colleagues have found to keep the critical period open longer.
Both excitatory and inhibitory neurons are important in establishing the critical period, with the inhibitory (GABA-secreting) cells serving as the trigger for flexibility in this system. The balance of activity between excitatory and inhibitory neurons is very much at the heart of the challenge to understanding epilepsy. Recent work suggests that changes that affect the critical period also change an individuals’ vulnerability to epileptiform activity. These results are important for autism because they undercover important functional anomalies in mouse models of autism that we may also expect to observe in some individuals with autism. For example in the Neuroligin 3 mutant mouse, the critical period for establishing eye dominance is completely absent. The competition for neural real estate, which is typically seen only very young animals, can be induced throughout the life of the Neuroligin 3 mouse. Similar atypical findings with respect to this early sensory flexibility or plasticity occurs in other animal models for autism. Lastly, but importantly, while the visual system’s plasticity may be well-established and produce easily observable effects, Dr. Fagilioni and her colleagues are examining the critical period in other sensory modalities like audition.
Considering these results together offers a greater appreciation of the complexities—and also the excitement—of investigations at the interface of epilepsy and autism. We need research investments at both ends of the discovery path. Greater investment in the basic research will help relates early sensory plasticity, and altered sensory experiences, to mechanisms of epilepsy and autism. We also need a better understanding of the drugs that can be made available now to deliver safe and effective treatment options for achieving optimal outcomes in individuals with autism. Autism Speaks is currently exploring some of theses issues in a partnership with the International League Against Epilepsy.