The phrase “you are what you eat” came to us through a bit of physiology snark. According to The Phrase Finder: “Anthelme Brillat-Savarin wrote, in Physiologie du Gout, ou Meditations de Gastronomie Transcendante, 1826: Dis-moi ce que tu manges, je te dirai ce que tu es. [Tell me what you eat and I will tell you what you are].” Not fair given the pain associated with gout but then it was not an easy time medically speaking.
Recently a report has come from Sahlgrenska Academy at Göteborg University in Sweden concerning the effects of the pregnant mother’s diet on the fate of the unborn child. The results, published in Pediatric Diabetes are quite strong. Mothers who ate vegetables every day had children who were less likely to develop type 1 diabetes than mothers who ate vegetables 3 times a week. Over 5,700 infants met the inclusion criteria for the study. Of these 191 (3.3%) had some evidence of pancreatic islet autoimmunity. Children were followed for several years and multiple blood samples were taken and analyzed for auto antibodies. Mothers filled out an extensive questionnaire outlining their diet during pregnancy including the frequency of vegetables, meat fish, pastries, bread, snacks, coffee, alcohol, etc. Very little was found to correlate with changes in the risk of the child developing diabetes later in life EXCEPT vegetable intake. Here the “p” value (the measure of the likelihood that the observation was due to chance) was .004. This means that the odds that the correlation was due to random chance was about 4/1000 (or 1 chance in 250). The average increase in risk due to decreased vegetable eating was about 1.7 fold. So the results are strong.
How might this work? In previous posts I discussed how the formation of the immune system is based on a combination of genetics and random chance (recombination). Food choice of the mother is an odd addition.
I am reminded of a fascinating set of reports concerning pregnant mothers who took folate supplements to decrease the risk of cleft palette in their infants. These children had an increased incidence of asthma. While quite different from type 1 diabetes, there is a connection. Both are autoimmune diseases. Now folate has a known metabolic property. It can contribute methyl groups. Methyl groups are simply –CH3: a carbon with 3 hydrogens. Its claim to chemical fame is that it is uncharged and by putting methyl groups on proteins or on DNA, for example, the local surface charge is altered, the shape of the molecule changes subtly, and interesting things begin to happen.
Currently there is a rather intense interest in how methyl groups affect the genome (our total collection of genes). We know that islands of methylated bases within the vast sea of DNA sequence can promote or inhibit the activity of gene expression. For this reason a group led by Dr. David Schwartz examined the methylation state of genomes before and after a high folate diet. They used mice for this purpose. First of all, like the human children the pups from these mice developed airway hyperresponsiveness. When their genomes were examined using some very new technology to examine the methylation state of the DNA bases it was found that certain genes associated with the immune system were methylated and that this changed their expression levels in the pups. So, it appeared that folate intake by the mother actually did create a genetic change in the infant that ultimately affected the immune system. The study of this sort of change in the regulation of genes where the actual DNA sequence does not change has a special name. It is called epigenetics. Could something similar be happening with vegetable intake?
Vegetables contain all sorts of interesting substances. They evolved these chemicals to keep us land creatures from eating them causing us, in turn, to evolve a complicated biotransformation system in our livers. Performing a search using the terms “vegetable” and “epigenetic” I found a study on phytoestrogens. These compounds are found in soybeans and when fed to pregnant mice, again a global change in DNA methylation was observed (among many other things tangential to this article).
Thus I would like to propose that the mechanism underlying the Swedish observation is ultimately an epigenetic one. The Schwartz group noted in their study that there seemed to be a critical window of fetal development, at least in the mouse, during which these sorts of changes to the genome were possible. Likewise I suspect the human fetus will have a critical window during which the genomic sensitivity to environmental perturbations will be greatest.
We are in really new terrain here. The Nature versus Nurture debate appears to be incorrect. Nature and nurture intertwine to create that which is us.