|Home | About | Journals | Submit | Contact Us | Français|
In the etiologic investigation of complex diseases and neuro-developmental disorders, the interaction of genetic factors and the environment have been evaluated by comparing disease risk as a function of environmental exposure in individuals who carry genetic variants of interest. While advances in technology have improved classification of genetic variant data, methods to ascertain environmental exposures remain largely underdeveloped. In this editorial, we discuss how mitotically heritable epigenetic marks in imprint regulatory elements may potentially serve as genome-wide biosensors of exposure, thereby reducing exposure misclassification, especially in studies evaluating the early origins of adult disease. Development of epigenetic biosensors for exposure monitoring will require a concerted effort in not only demonstrating that they are a stable record of past exposures, but also that the epigenetic marks are causally related to exposure.
Mapping the human genome has provided the foundation for studies investigating the role of genetic and environmental factors in the etiology of complex diseases and neurological disorders. Assessment of genetic risk factors is becoming increasingly comprehensive with the advent of high throughput, genome-wide approaches. The constancy of genotypes has enabled their use as risk factors in case-control and related hybrid study designs. These genetic risk factors not only carry a low misclassification potential, but they also promise to result in epidemiological studies focused on individuals at high risk of developing chronic complex disease.
Exposure ascertainment for complex diseases largely relies on participants to recall past lifestyle behaviors and diet. Poor recall, especially differential recall between cases and controls, remains a fundamental concern threatening the validity of epidemiologic studies. As we embark on studies evaluating exposure throughout human life, such as testing for the developmental origins of adult disease, [1-3] we can anticipate misclassification due to poor recall to worsen. A growing body of evidence now suggests that epigenetic features of the genome that regulate phenotype without changing the nucleotide sequence may provide a means of ‘recording’ past exposures,  and thus could be exploited to improve exposure assessment.
DNA methylation is the most-studied of the epigenetic modifications, owing to its heritable nature, stability, and ease of measurement. To determine the role of epigenetic changes in the etiology of diseases, gene promoters are most commonly screened for aberrant CpG methylation, and then evaluated with respect to outcome. This experimental approach is analogous to that used in genetic epidemiology studies. Unfortunately, promoter methylation can vary between tissues and as a function of life stage, [5-8] making it difficult to use promoter methylation as an epigenetic biosensor for environmental exposures.
In contrast, the imprint regulatory elements that result in parent-of-origin dependent monoallelic expression of imprinted genes are normally differentially methylated (i.e. one parental allele methylated and the other unmethylated),  providing a 50% methylation baseline in diploid cells. This provides an advantage over study of non-imprinted gene promoter or transposable element methylation profiles, where methylation alterations are unidirectional. The establishment of the parental allele-specific methylation patterns of imprint regulatory elements occurs very early in development, prior to germ layer specification.  Since DNA methylation patterns are faithfully transmitted during somatic cell division, the unique differential methylation pattern of the imprint regulatory elements is perpetuated throughout life in all tissues. Hence, methylation shifts at these differentially methylated regions (DMRs) have the potential to function as genome-wide, epigenetic biosensors for environmental exposures that either increase or decrease methylation, particularly during early development when they are most vulnerable to deregulation.
Individuals exposed to the Dutch famine prenatally have a higher incidence of chronic diseases, including a doubling in the incidence of schizophrenia, type-2 diabetes, coronary heart disease, hypercholesterolemia, and some cancers. [11-16] A recent study of the Dutch famine victims illustrates the utility of using DMRs in investigating the role of environmental exposures during gestation in the etiology of complex diseases later in life.  Exposure to severe caloric restriction periconceptionally was associated with hypomethylation of a DMR that controls IGF2  expression. Interestingly, this aberrant methylation was detected in peripheral blood specimens six decades post nutrient privation. Whether a concomitant change in IGF2 expression resulted in the enhanced incidence of chronic diseases observed in these individuals is presently unknown.
The evaluation of over-nutrition in the etiology of complex diseases could also benefit from the use of epigenetic biosensors. In the last decade, there has been an increase in the prevalence of folic acid intake worldwide.  Folate is a vitamin necessary as a source of carbon moieties used for nucleotide synthesis and DNA methylation. [19, 20] Periconceptional folate deficiency is associated with risk of neural tube defects in the offspring, prompting several countries, including the US, to fortify milled grain with 140ug of folic acid per 100 grams,  in addition to advising women of child-bearing age to supplement diets with folic acid. As a consequence, circulating folate levels among American women have doubled in the last decade. 
Animal experiments have shown that manipulating maternal folate levels permanently alters DNA methylation of transposable elements, leading to increases in obesity. [22, 23] It is unclear if similar modulations of the epigenetic profile occur in response to increased maternal folic acid intake in humans. What is known, however, is that case-control studies conducted in the 1980s suggested a lower risk of several cancers in folic acid users. More recent studies show either no association between folic acid exposure and cancer incidence  or enhanced risk of cancer formation since folic acid fortification began.[25, 26] While these findings may result from the dysregulation of gene expression, testing this possibility will be difficult unless an epigenetic measure of past exposure to over-nutrition is developed.
We briefly discussed the possibility of using imprint regulatory elements throughout the genome as epigenetic biosensors for improving exposure assessment in epidemiologic studies. This will undoubtedly require a more detailed understanding of environmental exposures that induce epigenetic alterations at these DMRs. It will also require fully defining the entire repertoire of human imprinted genes and their imprint regulatory elements – the imprintome. Without this critical information, our ability to diagnose, prevent, and treat chronic diseases and neurological disorders that plague us will remain compromised.
The authors acknowledge Joellen Schildkraut, Dora Ilyasova, Truls Ostbye and Robert Millikan for their comments on the text.
The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive license (or non exclusive for government employees) on a worldwide basis to the BMJ Publishing Group Ltd and its Licensees to permit this article (if accepted) to be published in JECH editions and any other BMJPGL products to exploit all subsidiary rights, as set out in our license (http://jech.bmj.com/ifora/licence.pdf).