• Sugar and Autism

    Todays post brought to you by Daft Punk’s Derezzed and Swedish House Mafia’s Greyhound.

    As I mentioned yesterday, my sister-in-law asked me about this research paper: A macroepigenetic approach to identify factors responsible for the autism epidemic in the United States

    Also as I mentioned, this is a curious paper.  It’s a review paper, not an experimental paper.  Now, that’s not a bad thing.  Sometimes, the best way to figure something out is to look at all the research and data that everyone has done and start comparing it.

    Epigenetics is an interesting area of study.  The idea is that environmental factors change how proteins are expressed.  Now, these changes are heritable and can affect several generations, but the nucleotide sequence is not altered.  The DNA strand may undergo a chemical change that changes the regulation of a gene or something like that.

    This review paper compares variations in diet and toxins to see if they can find the cause of the increase in cases of autism in the US.  Previous research identified a gene (PON1) whose varients appear to be associated with autism in the US, but not in Italy.  So, kids in both Italy and the US have this gene, but in the US a variant of the gene is highly correlated to autism, while the same gene variant is not correlated to autism in Italy.

    Another study attributed this difference to organophosphate pesticides.

    In conclusion, our genetic and biochemical findings appear compatible with a gene–environment interactive model of autism pathogenesis, whereby the genetic vulnerability component is conferred at least in part by ‘long’ GGC alleles at the RELN locus yielding reduced amounts of Reelin protein, and by R192 alleles at the PON1 locus possibly decreasing paraoxonase activity against OP compounds such as diazinon. The environmental component consists of subacute exposure to OPs during critical periods of prenatal neurodevelopment.

    (OP = Organophosphate Pesticide)

    That seems to be very likely to me.  These types of pesticides are very toxic.  They attack the nervous system of all animals (unlike BT, for example, that cannot harm mammals)  and have been implicated in everything from Alzheimer’s to ADHD.  If these chemicals are present in the US, but not in Italy, then that seems to be plausible to me.  OPs could be having an epigenetic effect… or even just an impact on development.

    But the authors of the macroepigentic approach paper disagree (and now it starts to get really odd).  They blame, get this, high fructose corn syrup.  They give a fair number of reasons, but there reasons all seem to be couched in “may” and “potential”.  While these words are common in science, in this case, I find them distressing.

    Basically, the authors of this paper have stated that a neurotoxin is less likely to be affecting fetal development and instead sugar is the most likely culprit.  That’s a pretty strong statement and, in my opinion, they just don’t have the data to back it up.

    Part of the problem is that it is very difficult to do controlled human trials.  Especially when diet and environmental toxins are concerned.  For example, back where I’m from, we would have mosquito planes fly over once a week during the summer.  Fogger trucks would drive through all the neighborhoods.  I still don’t know what they were using back then.  I know what they use now though.  But if a doctor asked me about chemical exposure when I was a kid, I still couldn’t tell him.  Of course, part of my problem is that I lived across the street from  chemical plant for half my life.

    The other issue I see is that HFCS isn’t anything but a slightly different form of table sugar.  Table sugar is sucrose, which is one molecule of fructose and one molecule of glucose.  They are easily disassociated in the intestines to be absorbed separately.  Now there are two kinds of HFCS that are commonly consumed by people in the US.  HFCS 55, usually used in soft drinks, is 55% fructose and 45% glucose.  HFCS 42, used in baked goods, cereals, other beverages, etc, is 42% fructose and 58% glucose.  Yes, the HFCS in processed food is less sweet than sugar.

    The authors of the macroepigenetics paper cited another article that I found curious.  Now, this report (cited by the macroepigentics paper) is stated as supporting this claim.

    With dietary zinc (Zn) loss and copper (Cu) gain from the consumption of high fructose corn syrup (HFCS) [16], metabolic processes required to eliminate heavy metals are impaired in children with autism [4].

    But the report itself actually says:

    The utilization of selected minerals when sugars were supplemented to basal diets
    was investigated in two separate, laboratory-controlled human feeding studies. Fructose-fed
    subjects had higher fecal excretions of iron and magnesium than did subjects fed sucrose.
    Apparent iron, magnesium, calcium, and zinc balances tended to be less positive during the
    fructose feeding period as compared to balances during the sucrose feeding period. Conversely,
    high fructose corn syrup (HFCS) did not affect the mineral balances when compared to
    sucrose feeding. Subjects fed fructose experienced diarrhea which possibly decreased absorption
    of minerals and thus increased fecal mineral losses. No such adverse effects were noticed when
    HFCS was fed.

    Now, note the difference (my emphasis).  Isn’t that curious?

    What the HFCS paper did show is that straight crystalline fructose did seem to cause diarrhea, which would change the absorption of minerals and metal ions.  But that’s it.  It wasn’t HFCS, it was fructose only.

    Based on this, I can only conclude that the link between HFCS and autism is tenuous at best and completely fabricated at worst.

    Another thing that I found odd,but not related to the studies was the HCFS/autism paper included a three paragraph section (and table) on the prevalence of autism and special education trends in the US.

    It really seems that the authors are pushing for causal links when the two systems (autism and HFCS) are only correlated.


    D’Amelio M, Ricci I, Sacco R, Liu X, D’Agruma L, Muscarella LA, Guarnieri V, Militerni R, Bravaccio C, Elia M, Schneider C, Melmed R, Trillo S, Pascucci T, Puglisi-Allegra S, Reichelt KL, Macciardi F, Holden JJA, Persico AM: Paraoxonase gene variants are associated with autism in north America, but not in italy: possible regional specificity in gene-environment interactions.

    Mol Psychiatry 2005, 10:1006-1016. PubMed Abstract | Publisher Full Text OpenURL


    Ivaturi R, Kies C: Mineral balances in humans as affected by fructose, high-fructose corn syrup, and sucrose.

    Plant Foods Hum Nutr 1992, 42:143-151. PubMed Abstract | Publisher Full Text OpenURL


    Dufault R, Lukiw W, et. al.: A macroepigenetic approach to identify factors responsible for the autism epidemic in the United States. Clinical Epigenetics 2012, 4:6  doi:10.1186/1868-7083-4-6

    Category: BiologyGeneticsScienceSkepticism


    Article by: Smilodon's Retreat