In 1965, Sir Austin Bradford Hill published nine “viewpoints” to help determine if observed epidemiologic associations are causal. Since then, the “Bradford Hill Criteria” have become the most frequently cited framework for causal inference in epidemiologic studies. However, when Hill published his causal guidelines—just 12 years after the double-helix model for DNA was first suggested and 25 years before the Human Genome Project began—disease causation was understood on a more elementary level than it is today. Advancements in genetics, molecular biology, toxicology, exposure science, and statistics have increased our analytical capabilities for exploring potential cause-and-effect relationships, and have resulted in a greater understanding of the complexity behind human disease onset and progression. These additional tools for causal inference necessitate a re-evaluation of how each Bradford Hill criterion should be interpreted when considering a variety of data types beyond classic epidemiology studies. Herein, we explore the implications of data integration on the interpretation and application of the criteria. Using examples of recently discovered exposure–response associations in human disease, we discuss novel ways by which researchers can apply and interpret the Bradford Hill criteria when considering data gathered using modern molecular techniques, such as epigenetics, biomarkers, mechanistic toxicology, and genotoxicology.

BackgroundData from epidemiological and animal model studies suggest that nutrition during pregnancy may affect the health status of subsequent generations. These transgenerational effects are now being explained by disruptions at the level of the epigenetic machinery. Besides in vitro environmental exposures, the possible impact on the reprogramming of methylation profiles at imprinted genes at a much earlier time point, such as during spermatogenesis or oogenesis, has not previously been considered. In this study, our aim was to determine associations between preconceptional obesity and DNA methylation profiles in the offspring, particularly at the differentially methylated regions (DMRs) of the imprinted Insulin-like Growth Factor 2 (IGF2) gene.MethodsWe examined DNA from umbilical cord blood leukocytes from 79 newborns, born between July 2005 and November 2006 at Duke University Hospital, Durham, NC. Their mothers participated in the Newborn Epigenetics Study (NEST) during pregnancy. Parental characteristics were obtained via standardized questionnaires and medical records. DNA methylation patterns at two DMRs were analyzed by bisulfite pyrosequencing; one DMR upstream of IGF2 (IGF2 DMR), and one DMR upstream of the neighboring H19 gene (H19 DMR). Multiple regression models were used to determine potential associations between the offspring's DNA methylation patterns and parental obesity before conception. Obesity was defined as body mass index (BMI) ≥30 kg/m2.ResultsHypomethylation at the IGF2 DMR was associated with paternal obesity. Even after adjusting for several maternal and newborn characteristics, we observed a persistent inverse association between DNA methylation in the offspring and paternal obesity (β-coefficient was -5.28, P = 0.003). At the H19 DMR, no significant associations were detected between methylation patterns and paternal obesity. Our data suggest an increase in DNA methylation at the IGF2 and H19 DMRs among newborns from obese mothers, but a larger study is warranted to further explore the potential effects of maternal obesity or lifestyle on the offspring's epigenome.ConclusionsWhile our small sample size is limited, our data indicate a preconceptional impact of paternal obesity on the reprogramming of imprint marks during spermatogenesis. Given the biological importance of imprinting fidelity, our study provides evidence for transgenerational effects of paternal obesity that may influence the offspring's future health status.

Through DNA methylation, histone modifications, and small regulatory RNAs the epigenome systematically controls gene expression during development--both in utero and throughout life. The epigenome is also a very reactionary system; its labile nature allows it to sense and respond to environmental perturbations to ensure survival during fetal growth. This pliability can lead to aberrant epigenetic modifications that persist into later life and induce numerous disease states. Endocrine disrupting compounds (EDCs) are ubiquitous chemicals that interfere with growth and development. Several EDCs also interfere with epigenetic programming. The investigation of the epigenotoxic effects of bisphenol A (BPA), an EDC used in the production of plastics and resins, has further raised concern for the impact of EDCs on the epigenome. Using the Agouti viable yellow (A vy ) mouse model, dietary BPA exposure was shown to hypomethylate both the A vy and the Cabp IAP metastable epialleles. This hypomethylating effect was counteracted with dietary supplementation of methyl donors or genistein. These results are consistent with reports of BPA and other EDCs causing epigenetic effects. Epigenotoxicity could lead to numerous developmental, metabolic, and behavioral disorders in exposed populations. The heritable nature of epigenetic changes also increases the risk for transgenerational inheritance of phenotypes. Thus, epigenotoxicity must be considered when assessing these compounds for safety.

No abstract