Sections

Services

Information

UK United Kingdom

Can a gut bacteria imbalance really cause autism?

ABC’s Four Corners recently aired a controversial documentary (The Autism Enigma) on how gut bacteria might affect autism. Here we review some of the science surrounding what it suggested caused autism…

The Autism Enigma presents one of many theories about the causes of autism. hepingting/Flickr

ABC’s Four Corners recently aired a controversial documentary (The Autism Enigma) on how gut bacteria might affect autism. Here we review some of the science surrounding what it suggested caused autism - the involvement of gut (or enteric) bacteria.

The human microbiome includes viruses, fungi and bacteria, their genes and their environmental interactions, and is known to influence human physiology. To suggest, as The Autism Enigma does, that the microbiome is altered in autism and that a prescription of antibiotics to modulate gut bacteria can “cure” autistic characteristics is premature.

Our knowledge of the microbiome is in its early stages, and two recent landmark studies (here and here) show that it is so complex that scientists can’t yet match gut flora profiles between healthy subjects.

There’s very broad variation in these bacteria in different people and that severely limits our ability to create a “normal” microflora profile for comparison among healthy people and those with any kind of health issue, including autism.

How the gut works

Let’s step aside from the puzzles of the microbiome for a moment to consider the basics of how our digestive tract functions. It may surprise some to hear it referred to as a mini brain, but our gut contains 500 million neurons and can go about its business independently of the brain, so the title is apt.

These enteric (gut) neurons are arranged in two web-like layers (plexuses) sandwiched between the two tubes of muscle that form the digestive tract. The neurons control the gut, contracting the muscle to propel or mix its contents, as well as triggering secretion of water and enzymes to aid digestion and absorb nutrients.

Although it’s too early to suggest that the microbiome is where the cause of autism can be found, some of the issues covered in the documentary are solidly supported by scientific literature. Gut bacteria interact with enteric neurons and the mucous membrane of the digestive tract, for instance, and changes in these bacteria can alter anxiety-like behaviours and memory function.

Signals produced by the metabolic products of the gut flora can indeed travel to the brain via several different routes, including via vagal and spinal sensory nerves, and alter function in the central nervous system. And changes in brain gene expression have been observed in mice fed some bacterial strains.

hepingting/Flickr

The role of fatty acids

Gut bacteria produce many products (including fatty acids) that affect neuronal function. One of the researchers interviewed in The Autism Enigma showed that rats became less social when their brains were injected with one particular fatty acid (propionic acid).

The documentary suggested that some fatty acids may cause autism; in particular regressive autism, where children forget skills learnt during their early years. It suggested that this might happen when fatty acid levels are altered due to dietary changes when families migrate to a new country and adapt to new foods. The specific example it provided was of Somali families with a traditionally high-fibre diets that migrate to Canada and adapt to a Western high-sugar diet.

The Autism Enigma specifically focused on propionic acid, a short-chain fatty acid product of bacterial fermentation in the gut. This is one of the three short-chain fatty acids (the others are acetic and butyric acids) that are the main fermentation products of bacteria in the gut. The main fuel for this fermentation process is dietary fibre.

Short-chain fatty acids have several roles that promote gastrointestinal health. They are the predominant energy source for the mucous membrane of the colon and important for secretion and absorption. And they maintain an acidic environment in the colon, which reduces the growth of pathogenic organisms.

Blaming sugar

The researcher studying rat behaviour suggested that excessive sugars in the diet contribute to increased propionic acid and gut problems. But given that dietary fibre is processed by the metabolic pathway that produces propionic acid and that common dietary sugars are digested and absorbed nearly three metres away in the upper small intestine, singling out a high-sugar diet as the culprit may be a simplification.

Gut bacteria are largely confined to the third portion of the small intestine, the pouch connected to the junction of the small and large intestines (caecum) and the adjacent portion of the colon. Sugars, proteins and fats are digested well before food reaches the colon, and are largely in the form of fibre by the time they reach the colon, not carbohydrates as suggested in The Autism Enigma.

Increased levels of propionic acid production are likely to be due to increased dietary fibre, not sugars per se. So, increased propionic acid production would logically correspond with a high-fibre diet, rather than a high-sugar diet. This contrasts with the documentary, which suggests diets high in carbohydrates and sugars would feed the gut bacteria. In fact, the original diet of Somali families with high levels of autism includes many fermented foods, which would also be high in propionic acid.

There are many theories about the causes of autism, and a role for the microbiome is only one of them. Early intervention behavioural therapy is, for now, the only treatment scientifically shown to improve behavioural symptoms of autism.

Sign in to Favourite

Join the conversation

35 Comments sorted by

  1. Felix Mendosa

    Researcher

    To what extent though would an association between the 'microbiome' and autism account for identified changes in the incidence of autism by cohort ... for more information refer here:

    Keyes, K. M., E. Susser, et al. (2012). "Cohort effects explain the increase in autism diagnosis among children born from 1992 to 2003 in California." International Journal of Epidemiology 41(2): 495-503. http://ije.oxfordjournals.org/content/41/2/495.abstract

    report
  2. ben walter

    Resident

    I have only anecdotal evidence unfortunately, but I believe autism, and it's steady linear increase in the population is strongly related to, or on the same spectrum as another genetic disorder called Phenylketonuria (PKU). This disorder produces the same phenotype as autism, and is due to a dysfunction in the liver enzyme that processes phenylalanine into tyrosine.
    This disorder, I postulate, can be induced by an excessive ingestion of phenylalanine, rendering the hepatic enzyme saturated, and…

    Read more
    1. Ernest Bennett

      Mr (retired)

      In reply to ben walter

      There is some support for this hypothesis. When the tricky little gut bugs get to work on the amino acids, for example, in dietary proteins a whole raft of bioactive metabolites are possible. The intestinal microbiome has not been definitively characterised and they vary from subject to subject. What can emerge from this biochemical free-for-all is yet to be defined in any detail. And the fatty acids as well! Overlaying this is, of course, is the indiscriminately profligate use of antibiotics. Clostridium difficile as a recognised problem?

      And for the editor: "...a gut bacteriUM", please.

      report
    2. Sue Ieraci

      Public hospital clinician

      In reply to Ernest Bennett

      (The singular article "a" refers to the singular noun "imbalance", not the plural noun "bacteria")

      report
    3. Sue Ieraci

      Public hospital clinician

      In reply to ben walter

      Ben Walter - in what way can you say that PKU "produces the same phenotype as autism"?

      "Phenotype" describes the expression of genes in an individual in their environment. PKU is a specific inherited disorder (autosomal recessive) that affects one metabolic pathway, and can be controlled by avoiding foods containing phenylalanine. A specific gene has been identified. People inheriting this condition also have clinical features including developmental delay, and generalized hypopigmentation.

      "Autism" is a spectrum of developmental /behavioural disorders of highly variable function, without strict inheritance (though familial tendency),a nd without an identified deficiency in the relevant hydroxylase enzyme.

      You are right, though, the theory does sound "very cliched and a conspiracy theory."

      report
    4. ben walter

      Resident

      In reply to Sue Ieraci

      It is only a theory, but as I stated previously I believe that some of the phenotypical features expressed in children with autism are also expressed in PKU - developmental delay and ADHD. I am not saying they are the same thing, I am saying that there are two ways to have too much phenylalanine, not being able to break it down (PKU), or having too much of it in the first place (Dietary-induced autism).

      Phenylalanine is a large, neutral amino acid (LNAA). LNAAs compete for transport across the…

      Read more
    5. mark mc dougall

      educator

      In reply to ben walter

      extraordinarily important - to get beyond commercial interests and to fund sound research. Needful also for GM foods and for roundup residues.
      Artificial sweeteners of increasing effect(semi hormonal) are increasingly being introduced- one can hardly even buy a fruit juice without them, and they are written in about 6point script on the label so one needs a magnifying glass to see it.
      Asthmatic tendencies can also be exacurbated along these lines -excessive contractive forces - towards spasticism -not enough balancing - expansion - sleep. MSG types also increase this overstimulation of the nervous system in its contractive pole. - paralysis.

      report
    6. Sue Ieraci

      Public hospital clinician

      In reply to ben walter

      ben walter - wouldn't it be easy to test your theory by measuring serum phenylalanine?

      What evidence is there of " not have any research being funded into the issue because of the potential ramifications of the results."? Are you proposing that potential research is being blocked by softdrink manufacturers? If this were the case, wouldn't they be opposing labelling for PKU sufferers?

      There is much greater evidence of research resources being diverted to chase and disprove Wakefield's discredited MMR proposal. In the meantime, there is a lot of active research, including the US NIH investing $100 million to the Austism Centers of Excellence Program - funding nine grant recipients for five years. There is lots more information available from the Autism Science Foundation website.

      No need for conspiracy theories.

      report
    7. Sue Ieraci

      Public hospital clinician

      In reply to mark mc dougall

      Mark - the vast majority of asthmatics react to either viral infections or allergens - most commonly pet dander or airborne polles or spores. Relatively inert chemicals are much less likely to act as allergens.

      Monosodium glutamate is a naturally-occuring substance which is only harmful if someone has a sensitivity or drug interaction - otherwise, like Sodium Chloride, it's a naturally occurring flavour enhancer. Both can do some harm in excess or to people with specific medical conditions, but neither cause illness in an of themselves.

      report
  3. Linus Bowden

    management consultant

    I don't know anywhere near enough about the science to have a strong opinion one way or the other. But in 4 Corners' defence, even this novice got the message loud and clear that they were presenting only one - relatively new - perspective.

    I found it quite gripping, as I had never rated 'the gut' as such a possible significant player in broader medical issues, particularly neurocognitive issues. But remember all those years when people with stomach ulcers were packed off for psychiatric evaluation to treat the 'stress/anxiety' that was widely blamed for stomach ulcers. Turned out they were caused by bacteria in the gut, and cured with antibiotics.

    report
  4. Robert Tony Brklje
    Robert Tony Brklje is a Friend of The Conversation.

    retired

    Junk chemicals in junk food place a trail of pharmaceuticals left over in the water supplies, all entering the child through the mother in the most critical years could all have an impact.
    Taking autism as being a developmental malady, a some what minor disruption of cerebral function becomes exacerbated because it disrupt the development of social functions.
    It seems to like the sufferer locked in a drug addicted state where they must maintain certain thought and behavioural patterns to sustain high brain reward states and when this is disrupted lash out at those denying them their natural high.
    Like all drug addicts that states of withdrawal and saturation are going to play merry havoc with all the bodies metabolic systems, confusing issues of cause and affect.
    Always important to pay attention to the less pronounced variations like Asperger syndrome and even misdiagnosis of psychopathy, a good look at the full range of social developmental issues that express at an early age.

    report
  5. Sue Ieraci

    Public hospital clinician

    Thank you to the authors for an informative article, and for their appropriate caution in relation to the proposals put forward in the Autism Enigma.

    While it is no doubt true that the gut is replete with neural tissue, it is also important to distinguish the central from the peripheral nervous system, and the voluntary from the autonomic. There is a long distance, as well as a long gap in function, between regulation of gut contraction and conscious thought (cognition).

    Gut flora and dietary…

    Read more
    1. Linus Bowden

      management consultant

      In reply to Sue Ieraci

      I love the phrase "gut flora". Or should it be "gut fauna"? :)

      report
    2. Sue Ieraci

      Public hospital clinician

      In reply to Linus Bowden

      Perahps both, Linus, to incorporate fungi. Hence "microbiota", I guess.

      report
    3. Naomi Bishop

      Researcher & Lecturer: Cell Biology in Health & Disease

      In reply to Sue Ieraci

      Hi All,

      Microbial flora or 'normal flora' (of the human or animal body) are commonly used terms/jargon used by Microbiologists. There is a move to using the term 'microbiota' in recent years, but the historial terms are still widely used (and understood) by Microbiologists.
      :-)

      report
  6. Steve Brown

    logged in via email @yahoo.com.au

    "But given that dietary fibre is processed by the metabolic pathway that produces propionic acid and that common dietary sugars are digested and absorbed nearly three metres away in the upper small intestine, singling out a high-sugar diet as the culprit may be a simplification."

    I wonder if the authors are familiar with the problem described by Sidney V. Haas and Elaine Gotschall. They wrote about how sugar and other complex carbohydrates weren't actually being properly digested in some people…

    Read more
    1. Sue Ieraci

      Public hospital clinician

      In reply to Steve Brown

      Steve - the writing of Haas is mostly in the area of Coeliac disease - which affected Gotschall's daughter. This is a specific familial disease with an autoimmune component, which can be diagnosed with both blood tests (looking for autoantibody) and small intestine biopsy.

      It is relatively straightforward to apply these tests to see whether a person has celiac disease or not. Not all "gut problems" involve destruction of intestinal villi.

      report
    2. Steve Brown

      logged in via email @yahoo.com.au

      In reply to Sue Ieraci

      "Not all "gut problems" involve destruction of intestinal villi."

      Indeed. But damage to villi isn't exclusive to coeliac's is it?

      report
    3. Sue Ieraci

      Public hospital clinician

      In reply to Steve Brown

      No, there are various causes of mucosal atrophy, involvinig the villi, including starvation and critical illness.

      The specific cause is identified on mucosal biopsy. Coeliac disease, for example, shows a typical histological picture that improveson a gluten-free diet. There are also characteristic histological changes that occur in the mucosal atrophy associated with starvation.

      Histopathologists look for characteristic patters of abnormality in structure of cell numbers and configuration that correlate with different conditions.

      If you are interested in the histopathology of the gut mucosa, there are many academic papers or tutorials on-line.

      report
  7. Naomi Bishop

    Researcher & Lecturer: Cell Biology in Health & Disease

    To me, it is also important to bear in mind cause and effect.

    People with autism cope with a lot of stress on a day-to-day basis, they also often have restricted diets. Diet and stress can each lead to changes in the gut microflora. So it may not be gut microflora causing the autism, but rather autism-induced behaviours leading to the changes in gut bacteria.

    Synergism is also another possibility, as if changes in gut bacteria are causing, or triggered by gut inflammation, then this could increase things like general irritability and poor concentration, for example.

    Another thing that puzzled me with the documentary was why the propionic acid was being directly introduced into the brain, and not being provided orally in food, or via a tablet/capsule...any thoughts?

    report
    1. Sue Ieraci

      Public hospital clinician

      In reply to Naomi Bishop

      Does propionic acid cross the blood-brain barrier?

      If it does, then oral intake should reach the brain. If it doesn't, then how can a product of gut flora influence brain function?

      report
    1. Sue Ieraci

      Public hospital clinician

      In reply to Jonathan Maddox

      This guy's work is speculative rather than experimental.

      As the news item says, cerebral Toxoplasmosis is a know condition - particularly in the immunosuppressed such as untreated HIV patients. This is a type of meningitis where the infecting organism is protozoal rather than bacterial or viral.

      it is well-known that people with encephalitis or meningitis can have altered behaviour - it's called "encephalopathy" or "cerebral irritation". This is quite different to speculating that people who dress in a certain way might have hidden meningitis, as this person appears to speculate.

      Like all of these speculations, there are logical ways of testing them and putting them to rest if they are not supported by evidence.

      report
    2. Jonathan Maddox
      Jonathan Maddox is a Friend of The Conversation.

      Software Engineer

      In reply to Sue Ieraci

      I understand that work is being done with the rats, though I don't know any details of the research. For confirming or disproving the hypothesis in humans with such subtle changes as are described, I think you'd need very large samples (as with any useful statistical psychological work) to draw concrete conclusions.

      report
    3. Sue Ieraci

      Public hospital clinician

      In reply to Jonathan Maddox

      Jonathan, you are right that you would need very large sample to test for subtle changes, but before even starting, the theory needs to be biologically feasible.

      Flegr is an evolutionary biologist. He talks about "latent" toxo infection - which is not a clinical concept. Many people have been exposed to toxoplasmosis, but not infected (much like carrying bacteria that don't infect you). Someone who is carrying, for example, meningococcal organisms in their nose is not "latently infected" - they are carriers. An infection generally occurs in the immune-impaired, leading - in the case of toxo - to the formation of cysts.

      There is nothing wrong with specultive thinking, but, when applied to human health, the testing needs to use biologically feasible concepts that accord with what we know about how the body works.

      report
    4. Jason Ozolins

      logged in via LinkedIn

      In reply to Sue Ieraci

      Not sure if there's a translation issue between Flegr's native tongue and English around the phrase "latent infection". Isn't he talking about a chronic, asymptomatic infection - asymptomatic that is, given *what people have been thinking of as symptoms*? Chagas' disease can remain dormant for years before reactivating - do people call that a latent infection?

      Just chanced upon this article: "Endochin-like quinolones are highly efficacious against acute and latent experimental toxoplasmosis…

      Read more
  8. Steve Melnikoff

    Physicist at University of Melbourne

    Another point that does not seem to have been emphasised in this article (and possibly the 4corners program) is the importance of recognising autism, as currently diagnosed, to be a 'spectrum' condition (and to note we are all just 'data points on the spectrum'). As such, I think just like with cancer, questions of cause need to be explicitly framed in terms of 'subtype' in order to help avoid notions of a singular, universal 'cure' (whatever that means for autism).

    As a reference and an example…

    Read more
  9. Jon Brock

    ARC Australian Research Fellow in Cognitive Science at Macquarie University

    Thanks for a really informative post Elisa. It's well outside my area but I was sufficiently intrigued to track down some of the original research studies mentioned in The Autism Enigma. I've summarised them in a blogpost. Even to a non-expert it's pretty clear that the science is a lot messier than the program-makers would have us believe.

    http://crackingtheenigma.blogspot.com.au/2012/09/the-science-behind-autism-enigma.html

    report
  10. Sean Lamb

    Science Denier

    When a condition, like Autism, which causes distress for so many people creates so much interest, it creates a market for single simplistic answers.
    That creates a great responsibility on the media on how they report on this issue and, because the media often isn't qualified to judge significance of scientific opinions, on the researchers themselves.

    Arguably that hasn't been meet with this 4Corners documentary. But I see the same flaws in this recent press release from the University of Melbourne…

    Read more
    1. Chris Borthwick

      logged in via Facebook

      In reply to Sean Lamb

      Not to mention that a 70% prediction rate produces [just to make the calculations easier, say autism has a prevalence rate of 0.5% rather than the claimed one in 150] in a test population of 2,000 three false negatives (people who have autism but test negative) and six hundred false positives (people who don't have autism but test positive). That is, the chance of having autism if you test positive is 7/600, or 1.1%. Not helpful, really.

      report
  11. Chris Borthwick

    logged in via Facebook

    Early intervention behavioural therapy has a good success rate because any early intervention program would have a good success rate, and ABA is the only organised program out there.
    Any program would 'work' because it's basically impossible to diagnose autism that early, and every mistaken diagnosis that doesn't turn into autism counts as a cure. The earlier you get them, the more cures you have. (The recorded 'cure' rates in ABA have always been pretty dicey, and the entire Lovass program rests on 28 subjects, but that's another matter).

    report
  12. Jason Ozolins

    logged in via LinkedIn

    "Sugars, proteins and fats are digested well before food reaches the colon, and are largely in the form of fibre by the time they reach the colon, not carbohydrates as suggested in The Autism Enigma."

    Umm, what about lactose intolerant individuals who unwittingly drink milk?

    Or breastfed babies whose mothers unfortunately produced excessive lactose-rich foremilk? Plenty of lactose got to the wrong parts of my son's digestive tract. Many, many sleepless nights nursing a very windy, screaming…

    Read more
  13. Anthony Everson

    logged in via Facebook

    I have an son who at 8yrs old was diagnosed with mild adhd and odd. this was becoming a problem for him at school with the teacher sending home notes and phoning us every other week with problems. We decided to take him to the Read Clinic for evaluation and posiblle causes as to why he was being so disruptive, the above was their diagnoses. When we inquired as to forms of treatment or action which could be taken to aleviate his problems the clinicians instantly stated "drugs".....my wife and I where…

    Read more