Can your pregnancy diet 'make your child fat'?

Tuesday April 19 2011

“A mother’s diet during pregnancy can alter the DNA of her child and increase the risk of obesity,” reported BBC News.

The news story is based on a study that looked at maternal diet and how it might be associated with “epigenetic changes” in the offspring. Epigenetics is the study of how genes can be influenced by the environment, without their DNA sequence being directly changed.

Researchers asked women to fill out food questionnaires during pregnancy, and then measured the fat levels of their offspring when they were older. They then compared these findings with DNA samples taken from the children’s umbilical cords. This well conducted study found associations between maternal diet, the likelihood that the child would have more fat at the age of six or nine, and chemical changes to a region containing a specific gene.

However, the researchers highlight that their findings demonstrate associations only. They do not show that maternal diet during pregnancy caused these changes, or that the epigenetic changes caused the children to have more fat. Further research is needed into whether this is the case. No recommendations for diet during pregnancy can be made based on this research. A healthy diet is an important part of a healthy lifestyle at any time, but it is especially vital if you're pregnant or planning a pregnancy. See our pregnancy care planner for more advice.

Where did the story come from?

The study was carried out by researchers from The University of Southampton, the University of Auckland and the Singapore Institute for Clinical Sciences. Funding was provided by WellChild, the University of Southampton, The Medical Research Council and the National Institute for Health. It was published in the peer-reviewed medical journal Diabetes .

What kind of research was this?

This genetics study looked at “epigenetic” changes to the DNA taken from umbilical cords of newborns and related these to the mother’s diet. Epigenetics is the study of how the environment can affect the function of genes. Signals from the environment can cause chemicals to be attached to DNA. These epigenetic chemical changes do not change the basic structure of DNA, and a gene that has had epigenetic changes will still make the same protein, but these changes may affect when the gene is switched on and the amount of protein the gene makes.

The researchers were interested in factors that affect the risk of human obesity and metabolic disease. They say that genomic variations (differences in the gene DNA sequences between people) explain only a fraction of the risk of obesity. Besides the child’s diet after birth, they say there is increasing epidemiological evidence that the mother’s diet during pregnancy may affect the child’s development.

They also say that animal studies suggest that maternal diet during pregnancy may lead to epigenetic modifications that alter the body composition of the offspring in adulthood. However, as yet there is no direct evidence in humans that such epigenetic processes during pregnancy are involved in children's later likelihood of obesity and there has been considerable debate about whether these modifications are significant enough to affect the children's development.

The researchers measured one type of epigenetic change called DNA methylation. They wanted to see if these changes were associated with the foetal environment in the womb and, furthermore, whether they were associated with the child’s weight at the age of six or nine years.

What did the research involve?

The study involved women that had been recruited into two different study groups (or cohorts) in Southampton. One group, from the Princess Anne Hospital (PAH) study, was made up of Caucasian women aged over 16 and less than 17 weeks pregnant with a single baby. The other group, from the Southampton Women’s Survey (SWS) was made up of women between 20 and 34 years who were not pregnant when they were recruited, but were then followed if they did become pregnant. Women with diabetes or hormonally induced conceptions were excluded.

The women in the PAH group were given a food frequency questionnaire when they were 15 weeks pregnant. The researchers then contacted them when their offspring reached nine years of age, and asked them to attend a clinic for follow-up. Of these, 219 children attended a clinic to have their fat levels measured. A DNA sample from the umbilical cord was available for 78 of these children.

In the SWS group, 239 children had both umbilical cord DNA available and childhood fat measurements when they were six years old.

From the DNA samples, the researchers selected 78 candidate genes that could be subject to epigenetic changes. From a subsample of 15 children from the PAH cohort, they looked at which genes from the umbilical cord sample had methylation changes above a 5% level. They then looked at which of these methylated genes were associated with obesity at nine years of age, and focused on five of these genes that could plausibly be involved in fat regulation.

What were the basic results?

The researchers found that similar numbers of mothers smoked in the two cohorts (21-34%). The average age of the mothers in the PAH cohort was 28, and 31 in the SWS cohort. The average body mass index (BMI) of the mothers was 22.3 in the PAH cohort, and 24.3 in the SWS cohort (a BMI of above 25 is considered overweight).

In the PAH cohort, the methylation of two genes was associated with childhood fat mass at nine years of age. These were the retinoid X receptor–α (RXRA) and endothelial nitric oxide synthase (eNOS). The researchers calculated that gender and these neonatal epigenetic changes were associated with more than 25% of the childhood variation in fat levels.

Higher levels of methylation of RXRA, but not eNOS, were associated with a lower maternal carbohydrate intake in early pregnancy. Fat and protein intake had no effect.

The amount of methylation at sites on two other genes (PIK3CD and SOD) was associated with the infant’s birth size.

For the SWS cohort, data were available for epigenetic methylation of genes from the umbilical cord and for fat levels at the age of six. In this group, eNOS methylation did not show an association with increased fat levels, but there was a similar association between RXRA methylation and fat levels as seen in the PAH cohort.

Sequencing of the RXRA gene showed that there were no particular sequence trends that could account for the differences in methylation seen between individuals. This means that it is unlikely that the differences seen stemmed from genetic variations between individuals.

How did the researchers interpret the results?

The researchers say that “greater methylation on the RXRA gene was associated with greater fat levels in later childhood”. They say that epigenetic measures at birth may be used to identify children at risk of obesity. This, they say, could potentially lead to programmes to optimise the mother’s health and nutrition with the aim of long-term benefits for the offspring. However, further research looking at methylation measurements in early life and comparing them with those in later life would be needed to evaluate how feasible this would be.


This was well conducted preliminary research, showing an association between methylation of one gene and increased fat levels in children when they were six or nine years of age. However, it should be noted that this was a relatively small study and further follow-up is needed to see how strong the association is.

The researchers found an association between lower consumption of carbohydrates during early pregnancy and increased methylation of the RXRA gene. It is important to emphasise that these associations do not necessarily mean that the mother’s diet caused this effect, or that different methylation patterns on genes cause childhood fat retention.

The researchers also point out that, although the food questionnaire is a validated study tool, there can be inaccuracies in people’s dietary reporting.

The study did find an association between lower carbohydrate and the methylation of the gene. However, is it unknown whether the amount of carbohydrate the women ate was within a healthy range. The researchers also did not say which foods the women had consumed. As such, it is not possible to say from this study whether the mother’s diet was “poor”. Further research is needed to explore which food groups, if any, are associated with epigenetic changes if any dietary recommendations are to be made to pregnant women.

Lastly, this study did not assess whether it is possible to control weight gain in the child associated with epigenetic changes by changing the diet during pregnancy.

A healthy diet is an important part of a healthy lifestyle at any time, but it is especially vital if you're pregnant or planning a pregnancy. See our pregnancy care planner for more advice.

Analysis by Bazian
Edited by NHS Choices