Autism Speaks Official Blog

While the word “mitochondria” is not yet a household word, it should be.  Many people know that mitochondria are the “powerplants” in our cells.  But mitochondria do much more than make energy.  Our mitochondria sense the environment we live in and adjust the chemistry of our cells to adapt to each new environmental change—cold, heat, drought, flood, infection, radiation, day, night, winter, summer, labor, rest, sleep, feast, famine, fear, and love.  All of these things are sensed by our mitochondria, and the metabolism and behavior of our cells is adjusted accordingly.  Mitochondria have memory.  When you exercise, your mitochondria grow and get better at just the kind of exercise you expose them to.  When they are exposed to a particular diet, they get better at using just that kind of diet.  The health and healthy development of our brains and bodies depend on the health of our mitochondria.  All the air, food, and drink we take in each day are processed by our mitochondria; sampled and transformed to literally allow us to think, to move, to laugh and play, to grow up, have families, and to grow old with grace.

The Shoffner Paper and Immunomitochondrial Biology
The special role of mitochondria in response to infection has been brought into sharper relief in recent years.  A new field that I call “Immunomitochondrial Biology” (IMB) is emerging.  IMB is shining a bright new light at crossroads in science that were formerly dark and mysterious.  In particular, this emerging field is shining new light at the crossroads of mitochondrial disease and autism.

In a “Top 10” paper by Shoffner, et al.¹ published in 2009, the role of mitochondrial dysfunction as a risk factor for regression in a subgroup of children with autistic spectrum disorders (ASD) was recently highlighted. About 4% of children with ASD have definite mitochondrial disease^2,3 (5 of 120 = 4.2%). Shoffner and his colleagues selected 28 children like this, with both mitochondrial respiratory chain disease and ASD. They found that 17 of these 28 children (17/28= 61%) had a history of a neurodegenerative episode that eventually grew into the features of ASD.  These were called “autistic regressions”, although it should be noted that the features of ASD typically evolve and change over time after an inciting event⁴.  They may start within 2 weeks of an event like infection, but they usually continue to evolve and change over weeks to months following the acute reaction⁴. In 12 children (12/17= 71%) the regressions began within 2 weeks of a fever greater than 101˚F.  In 4 of the 12 (33%; or 4/28 = 14% of all the cases of ASD and mitochondrial disease), the fever occurred after routine vaccination.  In 8 of the 12 (67%), the fever occurred after an infection or was of unknown origin.  Fever in patients with mitochondrial disease can occur with a known infection or inflammatory reaction, or can be a “fever of unknown origin” (FUO), without any cause that can be identified.  Five of the 17 (5/17=29%) had no fever or documented infection.  Eleven of 28 children (11/28=39%) had developed ASD gradually, without a history of episodic regression.  These numbers are shown graphically in Figure 1.

Figure 1.  The Findings of Shoffner, et al¹ on Autistic Regression in Children with Mitochondrial Disease and Autism Spectrum Disorder (ASD). The current best figures for the point prevalence of ASD in the US (blue circle; 1 in 110)¹⁰, the risk of “definite” mitochondrial disease among children with ASD (4%)², and the birth risk of childhood forms of mitochondrial disease (purple circle; 1 in 4000)⁸ are represented as Venn diagrams at the top.  The Top 10 Shoffner paper studied 28 children within the intersection of these 2 important disease groups.  The numbers in purple boxes represent the numbers reported in the study.  “FUO” is fever of unknown origin.  The standard clinical definitions for “possible”, “probable”, and “definite” mitochondrial disease are referred to as the modified Walker criteria³.

Neurodegeneration and Mitochondrial Disease
Children with mitochondrial disease have been known to be at risk for neurodegeneration after infection for some time⁵.  The risk for neurodegeneration (regression) following infection in conventional mitochondrial disease without ASD was first reported by our group at UCSD ⁵. Interestingly, the rate of risk that we reported, was the same as that found by Shoffner, et al¹ in 2009.  In 2002, we reported that 60% of children with mitochondrial disease (18 of 30 = 60%; 95% CI = 41% to 77%) suffered neurodegenerative events (regressions), and 72% of these events (13/18= 72%; 95% CI= 47% to 90%) were associated with infections in the period of time within two weeks before the onset of regression. We found no difference in the number of infections each child had in a year, which can be 3-10 infections per year in the first few years of life in normal children⁶.  However, in stark contrast to healthy children, children with mitochondrial disease were at risk for life-threatening neurodegeneration with otherwise common types of childhood infections.  The regressions in conventional mitochondrial disease in our series did not cause autism.  However, they did cause severe setbacks in motor, language, or cognitive milestones that were sometimes permanent.

The Flare and the Fade Response
There are two common types of regressions that parents and doctors can learn to recognize.  They differ by their timing and symptoms after the triggering event, and they differ in the kinds of problems that follow.  For simplicity, I call them the “fade” and the “flare” response.  When children with the common forms of mitochondrial disease suffer a regression, it is most often a “fade” response.  The fade response is typically delayed for 2-10 days after a fever resolves⁵, similar to the time course found in Reye syndrome in the 1980s⁷.  Parents typically report that their child was getting better from their cold or flu, when suddenly, their consciousness fades.  The child can become difficult to fully awaken, or will stop walking, stop talking, stiffen or lose muscle tone, or have a seizure, or a stroke-like episode.  The fade response involves an energy failure, and can lead to a series of neurodegenerative events and even death over the next 2-3 months, or to a self-limited event like a stroke-like episode that gradually gets better.  In contrast, the kinds of regressions that lead to an autistic regression are more often the “flare” response, similar to that suffered by Hannah Poling and described in the scientific literature⁴.  A flare response typically occurs early, at the peak of the fever and inflammatory response, within 2-3 days of infection.  During a flare response, there is a high fever, often over 102˚F, and hyperirritablilty, crying, inconsolability, a disrupted sleep-wake cycle, and a refusal to walk in children who might otherwise appear to be physically able to walk, choosing rather to crawl⁴.  Following a flare response, there can be a gradual evolution of other problems from persistent GI problems and diarrhea, a gradual loss of language over 2-3 months, with the onset of repetitive movements, to gaze avoidance and social avoidance⁴.  It must be emphasized that a flare response is not simply a high fever, or even a dramatic reaction to a high fever, like a febrile seizure.  It is a multisystem inflammatory response that carries a risk of autistic regression in genetically susceptible children.  One of these genetically determined susceptibilities is mitochondrial disease.

Different Kinds of Mitochondrial Disease
There are over 300 different kinds of mitochondrial disease⁸.  Are there special types that are at risk for autistic regression, while others that are not?  I believe the answer is, “Yes”.  In the year 2000, we reported our experience with a mitochondrial DNA (mtDNA) mutation⁹ that caused a very severe form of classical mitochondrial disease in an older sister (Leigh syndrome) when it was present in high concentration (86% heteroplasmy), but caused autism in a younger brother when present at a lower concentration (61% heteroplasmy).   Despite having precisely the same mtDNA mutation, the diseases we found in these two children were very different, and their development was very different.  There was no simple “genotype-phenotype” correlation.  The brain MRI in the sister with Leigh syndrome was very abnormal.  The brain MRI in her brother with autism was completely normal.  The sister had slowed development and the gradual onset of problems, with high blood lactic acid levels, seizures, heart abnormalities, ataxia and involuntary movements.  Despite this, her language skills were nearly normal.  She did not have an early history of significant regression, or loss of a milestone after it was achieved, although this did come later.  In contrast, the brother was hyperactive, had gained a few words normally, then lost them all by age 2.  He had a significant autistic regression without an identifiable trigger. His blood and spinal fluid lactates were normal.  These facts should remind us that mitochondrial disease, autism, and ASD are each developmental neurologic disorders.  The same trigger given at a different time to two children with the same DNA mutations will have a completely different outcome.

Future Research
We do not yet know how to predict which children are at risk, but current evidence suggests that children with the milder, rarer forms of mitochondrial disease might be at greater risk of autistic regression than those with more severe forms of mitochondrial disease that cause multi-system disorders.  The Top 10 Shoffner paper provides a touchstone for many important new questions.  Can certain kinds of mitochondrial defects actually cause the fever, and not the other way around?  Which kinds of mitochondrial defects lead to rapid, high-grade fevers in response to infection or vaccination?  Which defects lead to a failed fever response, or to a low-grade fever, or to a reduced immune response to vaccination?  Further research, and careful observations by both parents and pediatricians will help reveal the next clues that will provide hope and better treatment for children touched by these challenging disorders.

References

1.      Shoffner, J., et al. Fever Plus Mitochondrial Disease Could Be Risk Factors for Autistic Regression. J Child Neurol (2009).

2.      Oliveira, G., et al. Epidemiology of autism spectrum disorder in Portugal: prevalence, clinical characterization, and medical conditions. Dev Med Child Neurol 49, 726-733 (2007).

3.      Bernier, F.P., et al. Diagnostic criteria for respiratory chain disorders in adults and children. Neurology 59, 1406-1411 (2002).

4.      Poling, J.S., Frye, R.E., Shoffner, J. & Zimmerman, A.W. Developmental regression and mitochondrial dysfunction in a child with autism. J Child Neurol 21, 170-172 (2006).

5.      Edmonds, J.L., et al. The otolaryngological manifestations of mitochondrial disease and the risk of neurodegeneration with infection. Arch Otolaryngol 128, 355-362 (2002).

6.      Fox, J.P., Hall, C.E., Cooney, M.K., Luce, R.E. & Kronmal, R.A. The Seattle virus watch. II. Objectives, study population and its observation, data processing and summary of illnesses. Am J Epidemiol 96, 270-285 (1972).

7.      Partin, J.C. Reye’s Syndrome. in Liver Disease in Children (ed. Suchy, F.) 653-671 (Mosby, St. Louis, 1994).

8.      Naviaux, R.K. Developing a systematic approach to the diagnosis and classification of mitochondrial disease. Mitochondrion 4, 351-361 (2004).

9.      Graf, W.D., et al. Autism associated with the mitochondrial DNA G8363A transfer RNA(Lys) mutation. J Child Neurol 15, 357-361 (2000).

10.    Prevalence of autism spectrum disorders – Autism and Developmental Disabilities Monitoring Network, United States, 2006. MMWR Surveill Summ 58, 1-20 (2009).