Home > News & Opinion > News > 2012 > Dietary supplement may treat rare form of autism

Dietary supplement may treat rare form of autism

Popularity tracker
Emily Singer
6 September 2012

Simple source: By studying families in which the parents are related, researchers have found an inherited form of autism that might be treatable.

Researchers have uncovered a rare, genetic form of autism caused by mutations that speed up the breakdown of certain amino acids. The findings, published 6 September in Science, suggest that nutritional supplements may alleviate symptoms of the disorder in people with these mutations1

The mutations are in a gene called branched-chain ketoacid dehydrogenase kinase (BCKDK), which produces an enzyme of the same name. The enzyme inactivates the molecular machinery that typically breaks down branched-chain amino acids, a set of molecules that are essential for building proteins.

Mutant mice lacking the gene show symptoms similar to those seen in people with this rare form of autism — seizures and low levels of these amino acids in the blood. Feeding the animals a diet high in the amino acids reverses the symptoms within a week.

“As far as I know, this is the first example of a form of autism where there is a personalized treatment,” says lead investigator Joseph Gleeson, professor of neurosciences at the University of California, San Diego. “It [is] remarkable.”

The researchers have started treating individuals with the supplements, carefully titrating the doses so that blood levels of the amino acids resemble those in people with normal metabolism. They have been able to achieve near-normal levels of the amino acids, but they don’t yet know if symptoms improve.

Given the dearth of treatment options for autism — the only two drugs approved for the disorder are for irritability — the possibility of a simple treatment is significant.

“It’s exciting because this type of autism, at least, might be treatable with a metabolic intervention,” says Valerie Hu, professor of biochemistry and molecular biology at George Washington University in Washington, D.C., who was not involved in the study.

Mutations in BCKDK are rare — identified in just three families to date — so it’s unclear whether the results are relevant to other forms of autism. But the study adds to other evidence linking metabolic issues to the disorder.

Research published in May suggests that a genetic defect in the synthesis of carnitine, a molecule that plays a central role in the body’s ability to make energy, might slightly increase the risk of autism.  

“These are both inborn errors of metabolism that are linked to autism, which had not yet been identified because they don’t jump out on standard metabolic tests,” says Arthur Beaudet, professor of human and molecular genetics at Baylor College of Medicine in Houston, Texas, who led the carnitine research. “In both cases, there is concern about whether certain nutrients reaching the brain might be part of the problem.”

Essential aminos:

Amino acids are the building blocks of proteins. Branched-chain amino acids are the only ones the human body can’t synthesize and must be consumed in the diet. When branched-chain amino acids are in short supply, such as during periods of starvation, the BCKDK enzyme blocks their breakdown.

Gleeson and his collaborators identified mutations in this enzyme in two families, each with two children who have both autism and epilepsy. Both families are consanguineous, meaning the parents are related. The researchers found just one additional case of the disorder after searching their database of genomic data from consanguineous families.

The children with the mutations have lower levels of the branched-chain amino acids than their unaffected family members do, though their level of other amino acids is normal. “Their bodies handle branched-chain amino acids very differently,” says Gleeson.

It’s not yet clear how lack of these molecules leads to autism symptoms. But researchers speculate that the root cause lies in their transport into the brain.

Because amino acids cannot flow freely across the blood-brain barrier, many amino acids compete for the same molecular transporter. Lower levels of branched-chain amino acids in the blood means there are fewer to compete for transport into the brain.

Analysis of brain tissue from mutant mice supports this idea. These mice have lower concentrations of the branched-chain amino acids and higher levels of other amino acids in their brains than controls do. Either of these anomalies might cause seizures and autism, the researchers say.

The mutations don’t seem to affect neuronal shape or activity, however, suggesting that the mechanism at work is outside the neuron. “Our view is that the neurons are innocent bystanders,” says Gleeson.

Though mutations in BCKDK appear to be rare, it’s possible that mutations in other molecules involved in the metabolism or transport of branched-chain amino acids might have a similar effect.

For example, Hu’s research shows that expression of BCAT, a gene that encodes an enzyme that can either synthesize or break down branched-chain amino acids, is elevated in cell lines created from people who have a similar profile of epilepsy, autism and language impairment.

“Maybe the elevated expression is a compensatory mechanism to try to replenish some of the branched-chain amino acids,” she says.

The findings speak to the value of looking at genes that aren’t directly linked to neuronal function but still affect the brain, Hu says.

They also suggest a means of preventing this rare disorder, notes Beaudet. “One would like to identify children like this before they become symptomatic, with the idea that dietary intervention would prevent the disorder.”


1: Novarino G. et al. Science Epub ahead of print (2012) Abstract


Name: RAJensen
10 September 2012 - 10:39AM

The BCKDK gene maps to chromosome 16P11.2. Doesn't SFARI have a large group of families (VIP) whose children are diagnosed with a 16p11.2 mutation? Has SFARI ever looked for specific mutations in the BCKDK gene in the 16p11.2 region in this group who might possibly benefit from this treatment?

http://www.ncbi.nlm.nih.gov/gene?term=(bckdk[gene]) AND (Homo sapiens[orgn]) AND alive[prop] NOT newentry[gene]&sort=weight

Name: Alan Packer
10 September 2012 - 2:59PM

Unfortunately, while BCKDK does indeed map to the quite large 16p11.2 chromosomal band, it is actually not contained in the much smaller ~600-kb region at 16p11.2 that is commonly deleted in individuals with autism and other neurodevelopmental disorders.

Name: Hannah
15 September 2012 - 12:44AM

Is it then possible to check levels of carnitine or is this cost prohibiitive?

Name: Marco Loretto
16 September 2012 - 1:27PM

Is it possible that this mutation can take place as the result of a toxin? The recent 23 & me study has failed to identify any common genetic mutation. Yet there are many to be found in kids with autism.

Name: Emily Singer
17 September 2012 - 1:27PM

Thanks for your comment. A significant portion of the genetic risk of autism is now thought to come from rare genetic variants, which would not be detectable on 23andMe tests.

Name: Russell Bancroft
21 September 2012 - 2:35PM

Given the data, the supplement mentioned may cause significant positive changes to children with the disorder but it doesn't really claim to have treated someone. Studies are yet to prove something since experiments are still ongoing.

Name: Julie Lachapelle
6 October 2012 - 12:22PM

I was unaware that 2 drugs have actually been approved for the treatment of irritability in autistics. Can someone tell me what they are? Much thanks

Name: Emily Singer
6 October 2012 - 6:40PM

The two drugs are risperidone and aripiprazole. Both are antipsychotics also used to treat schizophrenia.

Name: Anonymous
3 April 2014 - 10:10AM

These inborn errors of metabolism are NOT that rare. How can they claim their rarity when the general population isnt routinely screened for them? They have no idea their prevalence rates. My daughter was never tested for them or autism until she went off to college and started having serious health problems. Over the last 3 years we have been on
an unending odysee trying to figure out what's wrong. Her childhood diagnosis of ADHD, dyslexia and other attention/ social behavior issues were changed to previously
undiagnosed autism and recently had mtDNA and nucDNA testing due to increasing
debilitating physical symptoms: bone crushing fatigue, marked
cognitive decline, gluten and casein senitivity, charbohydrate intollerence, elevated TPO, low aldosterone, low potassium, POTS, hyperprolinemia, increased infections, to name a few. Her gene results indicate Catch 22 Syndrome and numerous enzyme mutations mo
st notably missing the PRODH gene and carnitine. She is 25 yrs old!! She graduated Cum
Laude with a BA in History! She had an IQ of 138! She was running 5 miles a day! She was
fashion-model beautiful! Now her IQ is 108, she can barely get out of bed any more, has a very difficult time interacting with people, has little short term memory but is completely aware her body and mind are self destructing. Her autism, cognitive decline and extreme anxiety are directly a result from increasing PROLINE in blood due to the missing genetic code for the enzyme Proline Oxidase. So the obvious solution is to replace the missing enzyme with medication/ dietary supplements. But no one makes Proline Oxidase! Even though it can be manufactured like other digestive enzymes, no one does! Even though the scientific research reports hyperprolinemia is a risk factor for developmental dysfunction and schizophrenia, no one is connecting the dots! By the way, our daughter is not the product of a consanguinous marriage; both sides of the family genealogy have been researched back before the Revolutionary War and we share NO decendants. Side note: there are now 6 first cousins to our daughter with medically diagnosed autism 18 yrs- 5yrs. old.

Add a Comment

You can add a comment by filling out the form below. Plain text formatting.

Question: What is 10 + 4 ?
Your answer: